nature of the field
The geometry and some characteristics of the magnetic field is very
similar to one that would be produced, if a large bar magnet (a dipole
magnet - having two poles: north and south, by convention) were located
in the center of the Earth, somewhat inclined (about 11 degrees at present)
from the rotational axis of the Earth.
The dashed black lines with arrows represnt the lines of force of the magnetic field (represent the flow of the magnetic flux). Note that they are are perpendicular to the Earth's surface at the magnetic pole and tangential to the surface at the magnetic equator. The angle the magnetic field lines make with surface of the Earth is the magnetic inclination. The magnetic inclination is 90 degrees at the magnetic poles and zero degrees at the magnetic equator. It can be seen from the figure above that the magnetic inclination varies with latitude. A special compass whose needle is free to pivot in a vertical plane will point straight down at the poles and be horizontal at the equator.
The figure to the left represents a view down onto the Earth in the polar region. The red cross gives the location of the magnetic pole and the grey circle is the geographic or rotaional pole.
The angle between the two dashed lines is the magnetic declination.
The magnetic declination measured at location B is zero degrees, while at location A, it is approximately 25 degrees.
The magnetic declination varies with longitude. In reality it
is not so simple, because geometry of the main geomagnetic field is distorted
by the solar wind.
Despite the superficial similarities, the Earth' s dipole field can not be that of a bar magnet. The temperature in the center of the Earth reaches a few thousand degrees Celsius. This is very far above the curie temperature of any known materials. The curie temperature is that temperature above which a material looses its permanent magnetism. If a common bar magnet is placed in an oven and heated to 550oC, it will be de-magnetised.
If the magnetic field of the Earth is not permanent, it must be one that is always being generated. While they are rather complicated, theoretical models have been developed of the core of the Earth as a self generating dynamo. Such models (see note at bottom of this page) can account for the features of the field we observe and are consistent with physical principles. Without going into details of these models, consider the following.
origin of the field
Recall that the best evidence we have indicates that the core of the Earth is made of the metals iron and nickel (good electrical conductors). The solid metallic inner core is surrounded by the liquid outer core. The temperature of the liquid Fe-Ni solution is more than high enough to ionize the atoms.
Just as there are oceanic and atmospheric currents driven partly by
the fast rotation of the Earth, there are likely also such currents in
the core. Additionaly if the temperture increases with depth, the
lower part of the outer core would be hotter than its upper layers.
As you know such thermally layered liquids are unstable and tend to convect
heat upward. Thus it seems almost certain that currents in the outer
core (likely rather complex) exist , and this flow of charged ions can
produce a magnetic field. (Modern instruments can actually
measure the magnetic field of your heart as blood - an electrolyte - is
pumped through it).
what can be learned by measuring the remnant magnetism of rocks?
Most igneous rocks contain small quantities of the magnetic mineral "magnetite" (Fe3O4). Many sedimentary rocks also contain this mineral which might in sediments be formed by bacteria. When magnetite in a volcanic rock cools below its curie temperature, it becomes magnetized by the strongest magnetic field it can "feel" - i.e. the geomagnetic field. That rock there after possesses a dipole magnetic field whose north and south poles are exactly the same as that of the Earth's magnetic field.
When we measure the magnetic fields of geologically old rocks we find that their poles do not coincide (have the same orientation with respect to geographic north and south) with those of the Earths magnetic field. What do we infer from this ? Has the positions of the Earth's magnet poles changed or "wandered" over geologic time, or have the rocks themselves, or the continents that hold the rocks moved? See the bottom of this page.
By measuring the magnetic inclination and declinations of rocks of known age we can determine the locations of their continents at that time. Complications can occur and interpretations of such data must be made with care and always with more than one rock analysis.
Another important discovery that has been made by studing the paleomagnetism (ancient magnetism) of large numbers of well dated rocks is that from time to time the polarity of the Earth's magnetic field reverses itself. For most of the last 690,000 years the field has been normal (meaning as it is today), but prior to that it was completely opposite in polarity. That would be like turning that giant bar magnet around 180 degrees, so that north magnetic pole became south magnetic pole - or perhaps like reversing the direction of the flow of electrons in the electric magnet. Forty such reversal have been well documented over the past 15 million years.
During periods when the geomagnetic field is normal, volcanic rocks (including the basalts that erupt at oceanic ridges to form new oceanic crust) become normally magnetised. During periods when the geomagnetic field is reversed, cooling volcanic rocks become reversely magnetised. While the geomagnetic field undergoes many reversals through geologic time, once cool, the rocks' magnetization is locked in, so they preserve the history of reverals. The pattern of magnetic reversals recorded in the basalts that make up th eocean floors has provided a critical test for the sea-floor-spreading hypothesis. You can read about the test here.
(note: a recent mathmatical model for a core was runned on a supercomputer long enough to simulate 40,000 years of the Earth's core not only produced most of the characteristics of today's field, but unexpectedly went through a reversal)