The magnetic field of Earth is shaped like the one you see in a toy bar magnet, but there is a very important difference. The toy magnet field is firmly fixed in the solid body of the magnet and does not change with time, unless you decide to melt the magnet with a blow torch! The Earth's field, however, changes in time. Not only does its strength change, but the direction it is pointing also changes.

Map makers have been aware that the direction of the magnetic field changes since the 1700's. Every few decades, they had to re-draw their maps of harbors and landmarks to record the new compass bearings for places of interest. Think about it. If you are on a ship navigating a harbor in a fog, a slight change in your compass heading can take you into a reef or a sandbar!

Geologists have also been keeping track of the wandering magnetic poles as well. Instead of using compasses, they can actually detect the minute fossil traces of Earth's magnetism in rocks. These rocks are dated to determine when they were formed. From this information, geologists can figure out exactly how Earth's magnetic field has changed during the last two billion years. The results are surprising. Right now, the North point of your compass points towards the magnetic pole in the Northern Hemisphere. That's why compass creators put the 'N' on the tip of the magnetized compass needle. But because opposite's attract, this means that the magnetic pole in the Northern Hemisphere is actually a south magnetic pole! That's because scientists named magnetic polarity after the geographic compass direction!

Since the 1800's, Earth's magnetic South Pole which lives in the Northern Hemisphere has wandered over 1100 kilometers. By the year 2030, the magnetic pole will actually be almost right on top of our geographic North Pole. Then in the next century, it will be in the northern reaches of Siberia! Scientists are excited, and a bit concerned, by the sudden dramatic change in the magnetic pole's location. They worry that something may be going on deep within the Earth to cause these changes, and they have seen this kind of thing happen before.

What geologists have discovered is that the magnetic poles of Earth don't just wander around a little, they actually flip-flop over time. About 800,000 years ago, the Earth's magnetic poles were opposite to the ones we have today. Back then, your compass in the Northern Hemisphere would point to Antarctica, because in the Northern Hemisphere the polarity had changed to 'North' and this would have repelled the North tip of your (magnetized) compass needle. Geologists have discovered in the dating of the rocks that the magnetism of Earth has reversed itself hundreds of times over the last billion years. Careful measurements of rock strata from around the world confirm these reversal events in the same layers, so they really are global events, not just local ones. What is even more interesting is that the time between these magnetic reversals, and how long they last, has changed dramatically. 70 million years ago, when dinosaurs still roamed the landscape, the time between magnetic reversals was about one million years. Each reversal lasted about 500,000 years. 20 million years ago, the time between reversals had shortened to about 330,000 years, and each reversal lasted 220,000 years.

Today, the time between reversals has declined to only about 200,000 years during the last few million years, and each reversal lasts about 100,000 years or so. When did the last reversal happen?

This is a plot of the change in the main field strength of Earth for the last 800,000 years from the research by Yohan Guyodo and Jean-Pierre Valet at the Instuitute de Physique in Paris published in the journal Nature on May 20, 1999 (page 249-252).

The Brunhes-Matuyama Reversal ended 980,000 years ago when the polarity of the field actually did 'flip'. Since that time, the polarity of Earth's field has remained the same as what we measure today with the Northern Hemisphere Arctic Region containing a 'South-Type' magnetic polarity, and the Antarctic Region containing a 'North-type' polarity. You will note that the last reversal ended when the magnetic intensity reached near-zero levels. Since then, there was a near-reversal about 200,000 years ago labeled 'Jamaica/Pringle Falls' after the geologic stratum in which these intensity measurements were first identified. Scientists do not know just how low our field has to fall in intensity before a reversal is triggered, but the threshold seems to be below 2.0 units on the scale of the above 'VADM' plot. Beginning in the 1920's, geologists discovered traces of the last few magnetic reversals in rock samples from around the world. Between 730,000 years ago to today, we have had the current magnetic conditions where the South-type magnetic polarity is located in the Northern Hemisphere near the Arctic. Geologists call this the Brunhes Chron. Between 730,000 to 1,670,000 years ago, Earth's magnetic poles were reversed during what geologists call the Matuyama Chron. This means that the North-type magnetic polarity was found in the Northern Hemisphere. Notice that the time since the last reversal (the end of the Mayuyama Chron) is 730,000 years. This is a LOT longer than the 200,000 years!

Some scientists think that we may be overdue for a magnetic reversal by about 500,000 years!

Is there any evidence that we are headed towards this condition? Scientists think that the sudden, rapid change in our magnetic pole location is one sign of a significant change beginning to occur. Another sign is the actual strength of Earth's magnetic field.

Scientists are convinced that Earth's magnetic field is created by currents flowing in the liquid outer core of Earth. Like the current that flows to create an electromagnet, Earth's currents can change in time causing the field to increase and decrease in intensity. Geological evidence shows that Earth's field used to be twice as strong 1.5 billion years ago as it is today, but like the weather it has gone through many complicated ups and downs that scientists don't have a real good explanation for, or ability to predict. But the fossil evidence does tell us something important.

In the 730,000 years since the last magnetic reversal, Earth's field has at times been as little as 1/6 its current strength. This happened about 200,000 years ago. Also, around 700 AD it was 50% stronger than it is today. There have been many sudden ups and downs in this intensity, but some scientists think that conditions are rapidly becoming very different than the past historical trends have shown.

We've only been able to measure the Earth's magnetic field strength for about two centuries. During this time, there has been a gradual decline in the field strength. In recent years, the rate of decline seems to be accelerating. In the last 150 years, the strength of Earth's field has decreased by 5% per century. This doesn't seem like a very fast decrease, but it is one of the fastest ones that has been verified in the 800,000 year magnetic record we now have. At this rate, in 10 centuries we will be 50% below our current field strength, and after 2000 years we could be at zero-strength. The data on past reversals seems to show that, when the field reaches 10% of its current strength, a magnetic reversal can be triggered. It has been 730,000 years since the last reversal ended. We are certainly long overdue for a reversal, by some statistical estimates.

But the caveat is that magnetic changes come in a variety of timescales from the major reversal events every few hundred thousand years to micro changes called 'excursions' that come and go withing a few thousand years. Two detailed "studies of the geomagnetic field in the last 1 million years have found 14 excursions, large changes in direction lasting 5-10 thousand years each, six of which are established as global phenomena by correlation between different sites. Excursions appear to be a frequent and intrinsic part of the (paleomagnetic) secular variation".(Gubbins, David. 1999. The distinction between geomagnetic excursions and reversals. Geophysical Journal International, Vol. 137, pp. F1-F3.). The figure below shows on the left-side the magnetic intensity measurements since 500,000 years ago during the current Brunhes magnetic chron. You can easily see the 'spiky' fast excursions, but the overall magnetic intensity is decreasing in time to the present day. We may be living inside one of these fast excursions which will be replaced by a growing field in a few thousand years, but it seems that the big picture is still that the overall largescale field is declining slowly over 100,000 year timescles. It isn't the excursions we need to worry about for 'reversals' but this larger trend downwards that seems to be going on.

So, what will happen when the field reverses? The fossil record, and other geological records, seem to say 'Not much!'

Scientists have recovered deep-sea sediment cores from the bottom of the ocean. These sediments record the abundance of oxygen atoms and their most common isotope: Oxygen-18. The increases and decreases in this oxygen isotope track the ebb and flow of periods of global glaciation. What we see is that, during the time when the last reversal happened, there was no obvious change in the glacial conditions or in the way that the conditions came and went. So, at least for the last reversal, there was no obvious change in Earth's temperature other than what geologists see from the 'normal' pattern of glaciation. By the way, because glaciation depends on the tilt of Earth's spin axis, this also means that a magnetic reversal doesn't change the spinning Earth in any measurable way.

Loess deposits in China have recently given climatologists a nearly unbroken, continuous record of climate changes during the last 1,200,000 years. What they found was that the sedimentation record shows the summer monsoons and how severe they are. The only significant variation in the data could be attributed to the coming and going of glacial and inter-glacial periods. So, summer monsoons in China were not affected by the reversal in any way that can be obviously seen in the climate-related data from this period. The fossil record, at least for large animals and plants, is even less spectacular when it comes to seeing changes that can be tied to the magnetic reversal.

The Brunhes-Matuyama reversal happened 730,000 years ago during what paleontologists call the Middle Pleistocene Era (100,000 to 1 million years ago). There were no major changes in plant and animal life during this time, so the magnetic reversal did not lead to planet-wide extinctions, or other calamities that would have impacted existing life. It seems that the biggest stresses to plant and animal life were the comings and goings of the many Pleistocene Ice Ages. This led very rapidly to the evolution of cold-tolerant life forms like Woolly Mammoths, for example.

So, it seems that we may be headed for another magnetic reversal event in perhaps the next few thousand years. This event, based on past fossil and geological history, will not cause planet-wide catastrophies. The biosphere will not become extinct. Radiation from space will not cause horrible mutations everywhere. Ocean tides will not devastate coastal regions, and there will certainly not be volcanic activity that leads to global warming.

Of course, scientists cannot predict which minor effects may take place. A magnetic reversal could be a big nuisance to many organisms that will not lead to their extinction, but it just might lead to temporary changes in the way they would normally conduct themselves. The fossil record doesn't record how a species reacted to minor nuisances! Some animals use Earth's field to magnetically navigate, but we know that these same animals have back-up navigation systems too. Pigeons use Earth's magnetism to navigate, as do dolphins, whales and some insects. They also use their eyes as a backup, and a knowledge of land forms and geography, or the location of the Sun and Moon to get about. Humans have used compasses to navigate for thousands of years, but now we rely almost entirely on satellites to steer by. In the future, only those few anachronistic people using the ancient technology of compasses to get around, would have any problems!

The magnetic field of Earth shields us from cosmic rays, so losing this shield may seem like a big deal, but it really isn't. Cosmic rays are not the same kind of radiation as light, instead it consists of fast-moving particles of matter such as electrons, protons and the nuclei of some atoms. Our atmosphere is actually a far better shield of cosmic radiation than Earth's magnetic field. Losing the magnetic field during a reversal would only increase our natural radiation background exposure on the ground by a small amount - perhaps not more than 10%. The long term result might be a few thousand additional cases of cancer every year, but certainly not the extinction of the human race.

Return to Dr. Odenwald's FAQ page at the Astronomy Cafe Blog. References: Guo, Zhengtang, et al., 2000, "Summer Monsoon Variations Over the Last 1.2 Million Years from the Weathering of Loess-soil Sequences in China", Geophysical Research Letters, June 15, pp. 1751-1754. Guyodo, Yohan and Valet, Jean-Pierre 2003, "Global Changes in Intensity of the Earth's Magnetic Field During the Past 800 kyr", Nature, May 20, 2003, p. 249. Jacobs, J. A., "Reversals of the Earth's Magnetic Field, (pp. 48-50) Jacobs, J. A. "Geomagnetism" Academic Press (pp. 186-89, 215-220, 236-42) Merrill, Ronald, McElhinny, M. and McFadden, P., "The Magnetic Field of the Earth", Academic Press, (pp.120-125) Raymo, M., Oppo, D. W., and Curry, W. 1997, "The Mid-Pleistocene Climate Transition: A deep Sea Carbon Isotopic Perspective", Paleoceanography, August 1997, pp. 546-559. Rikitake, Tsuneji and Honkura, Yoshimori, "Solid Earth Geomagnetism", D. Reidel Publishing Co. (pp. 42-45) Ruddiman, W., et al. 1989, "Pleistocene Evolution: Northern Hemisphere Ice Sheets and North Atlantic Ocean", Paleoceanography, August, pp. 353-412. Wollin, G., Ericson, D., Ryan, W. and Foster, J. 1971, "Magnetism of the Earth and Climate Changes", Earth and Planetary Science Letters, vol. 12, pp. 175-183.