Geoscience Reference
In-Depth Information
Taking the Earth's temperature
The deeper you go, the hotter it gets, but how hot is the middle of the Earth? The answer
is that, at the boundary between the molten outer core and the solid inner core of the Earth,
the temperature must be at exactly the melting point of iron. But the melting point of iron
under those incredible pressures will be very different from its value on the surface of the
Earth. To find out what it is, scientists must recreate those conditions in their laboratories or
calculate them from theory. They've tried two different practical methods: one using tiny
samples squeezed between diamond anvils, the other using a giant multi-stage compressed
gas gun to compress samples just for an instant. Because of the difficulties in achieving
such incredible pressures - 330 gigapascals at the inner core boundary - and because of
the difficulty in calibrating pressure so that you know when you've got there, both meth-
ods have yet to measure that temperature directly. What they can do is measure the melting
point of iron at slightly lower pressures and try to extrapolate downwards. But there are
still difficulties. Not least because the core is not pure iron and impurities can affect the
melting point. Theoretical calculations put the inner core boundary at about 6,500 degrees
Celsius for pure iron, maybe 5,100-5,500 degrees Celsius for iron with the probable range
of impurities in the core. These are between estimates from diamond anvil and gas gun ex-
periments.
The study of seismic waves passing through the inner core has produced one more surprise.
The waves seem to travel 3-4% faster through the inner core when going north to south
compared to east to west; the inner core exhibits anisotropy, a structure or grain which is
not the same in all directions. The explanation could be that the inner core is made up of
lots of aligned crystals of iron - or even one big crystal more than 2,000 kilometres across!
It's also possible that there could be convection within the inner core just as there is in the
mantle. And there may be a small amount of liquid caught up in the crystalline mush. It has
been calculated that between 3% and 10% by volume of flat discs of liquid aligned with
the equator would give the inner core the anisotropy observed.
Spinning core
Like the Earth as a whole, the inner core is rotating, but not exactly in the same way as
the rest of the Earth. It is in fact rotating slightly faster than the remainder of the planet,
gaining nearly one-tenth of a turn in the past 30 years. Careful study of seismic waves from
earthquakes in the South Sandwich Islands off the southern tip of South America that were
detected in Alaska show the effect. It is revealed due to the north-south anisotropy in the
inner core that we just discussed. As the inner core pulls ahead of the rest of the Earth, the
effect due to that anisotropy changes. Seismic waves that skimmed past just outside the in-
ner core arrived in Alaska just as quickly in 1995 as they did in 1967. But waves passing
