Anyone who has watched a needle spin around a compass, or the colorful northern lights grace the skies near Earth’s poles, has seen the effects of Earth’s magnetic field. Researchers do not fully understand, however, the processes governing the field’s behavior. New information gleaned from old ships’ logs is now leading to a better understanding of the magnetic field’s past strength, which is turning out to be more erratic than some scientists previously thought.

Starting in about 1840, scientists such as mathematician Carl Gauss first established observatories to make precise measurements of Earth’s magnetic field. Since that year, researchers have recorded a 5 percent decrease per century in the field’s strength. Because of the lack of data prior to 1840, however, scientists assumed a similar decreasing trend for the field before that year.

To find out more about the field before 1840, David Gubbins, an earth scientist at the University of Leeds in the United Kingdom, and students looked to a database of magnetic information previously collected from old lava and archaeological remains (see Geotimes, October 2005). The database contains 315 measurements of magnetic intensity from between 1590 and 1840. Each measurement, however, has about a 10 percent error, which is as much as Earth’s magnetic field could change over a few hundred years, rendering single measurements “useless” in tracking field changes over small, century-long timescales, Gubbins says.

To work around the error problem, Gubbins and colleagues turned to directional measurements from compasses recorded in ships’ logs. The East India Company, for example, made routine compass measurements each day and by the 18th century, at least 50,000 such measurements existed worldwide, Gubbins says. The team combined the measurements of magnitude (from the database) and direction (from the ships’ logs) into a known equation and converted each of the 315 measurements into relative intensities. Those 315 scaled measurements allowed Gubbins to reduce the amount of error and “tie down” the rate of fall in the magnetic field.

Much to Gubbins’ team’s surprise, instead of the declining strength assumed before 1840 by most scientists, the field was essentially stagnant back to 1590, the earliest year included in the study, which the team published in the May 12 Science. The change was “effectively zero,” Gubbins says, “consistent with the field not changing at all in that time.”

The finding “differs from what had been a general conclusion,” that held that the field was decreasing centuries before 1840, says Bradford Clement, a geologist at Florida International University in Miami. But Gubbins’ paper “appears reasonable,” he says, and he is not surprised to see the erratic shift from stagnation to steady decline after 1840.

Indeed, the research appears “very sound,” and agrees with paleomagnetic measurements, says Cathy Constable, a geophysicist at Scripps Institution of Oceanography in La Jolla, Calif. When looking at Gubbins’ results in the context of the past 7,000 years, it becomes apparent that the current decrease is a “temporary feature,” and “not at all anomalous,” Constable says. Another model, for example, suggests three or four centuries-long intervals since about 500 B.C., during which the field actually increased in strength, she says.

The results then beg the question why, at about 1840, the field took a downward trend. “It’s the chicken and the egg,” Gubbins says. “The real cause is the dynamo inside the core, which we can’t see.” The dynamo, while not entirely understood, is the process by which hot, molten materials deep in Earth move to create the planet’s magnetic field.

The team could see, however, that the change from a stagnant to decreasing field at about 1840 corresponds to the onset of anomalous “patches” on Earth, where the magnetic field’s direction has reversed to become opposite of the field’s direction elsewhere on the globe. Directional measurements show that the patches first emerged by about 1700 near the equator, where they would not affect the magnetic field’s strength. By about 1800, however, patches of increasing size began to affect the field as they drifted south toward Antarctica — evidence supporting the idea that changes to the field started at about that time, Gubbins says.

Whether the field will continue on its downward trend and eventually completely reverse — so that compass needles point south instead of north — or whether it will level off or even increase, remains to be seen. Such magnetic flips have taken place multiple times in Earth’s history.

Kathryn Hansen

Links:
"Geophysics and archaeology collide," Geotimes, October 2005

Back to top