The Earth's magnetic field goes through periods of instability every so often. Sometimes the field comes back reversed (so that magnetic North reappears where the magnetic South used to be, and North where South was), but more often than not, the field goes through a weak period and then strengthens back to the way it was before. We do not completely understand the process, but we are close. It's one of those non-linear magneto-hydro-dynamic problems even your best professors had nightmares about. Even using our most powerful supercomputers, we have problems modeling this process accurately.
We are pretty sure this occurs on Earth, because when lava cools, the magnetic minerals preserve the direction of the magnetic field at the time. We can see that the direction preserved in progressively older rocks changes in the way described above: sometimes the field reverses completely, and sometimes the signal dimishes only to reappear again in the same direction.
This "magnetic pole reversal" pattern can be mapped out by looking at successively older and older rocks from different locations to get an idea of how often this change occurs. Here is what it looks like for the last 5 million years, with black being 'normal,' and white 'reversed':
(the above is from Lisa Tauxe's notes for her courses at UCSD)
There is actually more detail here than shown: very short-lasting reversals are omitted, so what we are looking at above is the 'dominant' polarity over 10- to 20,000 year periods. These dominant polarity periods, then, last 250,000 years on average, with a fair amount of variation. But what actually happens during a magnetic field reversal?
It takes about one thousand years or so for the field to reverse. At first, the field weakens, as is happening presently. The next part is interesting: the field can pass through a multi-pole phase. In other words, there is a crazy period where there are many, many North magnetic poles, and many, many South magnetic poles. Out of this chaos, the field slowly organizes again, and can emerge either 'normal' or 'reversed.'
What I have wondered about is the effects that these collapses/multi-pole phases might have had in the past, and what they will have in the future. Consider the timescales involved: one thousand years is quite fast compared to evolutionary (speciation) time scales, and fast even compared to things like migration or species expansion into biomes. For species that rely on the magnetic field for navigation (bees, tuna, some turtles, some birds, perhaps some whales) this kind of thing could be catastrophic if the field is their only cue. Are magnetic reversals associated with any particular die-offs? I have never seen any attempts to answer this question, but given the timescale precision required to date die-offs against magnetic reversals, as well as proving that the species in question had a magnetoreceptor, I am not surprised.
The other end of the question is that since there is no longer a strong magnetic field, much more of the Earth's surface is exposed to hard radiation. Think aurora borealis/australis scattered all over the place. Would this affect the speciation rate from radiation induced mutations? Again, a question that would be exceedingly difficult to approach, given the quality of the geologic record.
What about the present? What would happen to us today if we lost the magnetic field? There would be a higher radiation risk generally - but even especially so for airplane flight and for manned space flight. We could expect many satellite outages, and the accompanying data/communications problems. On the ground, electric grid disturbances would be stronger, and might cause more the failures similar to those we saw last year. There might be ozone holes all over the place, since the high energy rays that break apart ozone can only get in around the magnetic poles. We would probably all be able to see the aurora - even in equatorial areas. We might even see an increase in public interest in science.
Learn more.
