On March 11 of this year, an earthquake measuring 9 on the Richter scale produced a giant wave, or tsunami, that devastated the east coast of northern Japan, caused almost 16 thousand deaths and left roughly 10 thousand more missing or injured. In the midst of the news of the catastrophe, the press disclosed an interesting fact: according to US and Italian geophysicists, the Japanese earthquake shifted the axis around which the earth’s mass is distributed by a few centimeters. As a result of one tectonic plate sliding underneath another one during the tremor, the rearrangement in the planet’s mass also sped up the earth’s rotation and shortened the length of a day by 6.8 millionths of a second, producing an effect similar to that seen when an ice-skater begins to spin faster when he brings his arms in close to his body.
However, these two subtle geophysical changes are not the only ones produced by earthquakes followed by tsunamis. According to a study by researchers at the National Institute for Space Research (Inpe), in the city of São José dos Campos, and at the National Observatory (ON), in Rio de Janeiro, these major natural phenomenon cause infinitesimal disturbances in the earth’s magnetic field that can be measured and used to monitor the emergence and development of these giant waves. The feasibility of this approach is advocated in a scientific article that has just been submitted to an international journal. According to the Brazilian geophysicists, this study’s conclusion could form the basis of significant improvement of the current tsunami warning systems, and at a low cost.
It is common knowledge that the oceans can have a subtle effect on the magnetic field that is detected by compasses and generated at the center of the earth. Back in the late 1960s, researchers measured the variation in the geomagnetic field produced by the daily movements of the tides. The dissolved salt in the form of electrically charged chlorine and sodium ions makes seawater a conductive liquid of electricity. The movements of this liquid in relation to the earth’s magnetic field induce small electrical currents in the sea, explains the geophysicist Virgínia Klausner, from the UN, one of the authors of the tsunami study.
Known as the dynamo effect, this phenomenon is the same one that generates electric current in a piece of conductive wire when this is placed near a magnet, declares the physicist Odim Mendes Junior, from Inpe, one of Virgínia’s PhD advisors. “In turn, these sustained electrical currents in the sea create a magnetic field that overlaps the earth’s magnetic field and that can be measured with the proper magnetometers,”says Mendes, whose work is financed by FAPESP.
However, measuring the magnetism of a tsunami seemed impossible until very recently. Whereas the intensity of the earth’s magnetic field is of the order of 30 to 50 thousand nanoteslas – 20 times less than that of a fridge magnet – the variation in this field caused by a tsunami would fall between 1 and 10 nanoteslas. There are already magnetometers that are sufficiently precise to be able to measure these variations, but the signal can be masked by magnetic disturbances that are hundreds of times stronger, caused, for instance, by solar storms.
However, the sun was undergoing an exceptionally calm phase, when, on the February 27, 2010, an earthquake on the coast of Chile, measuring 8.8 on the Richter scale, led to a tsunami that spread across the entire Pacific Ocean. With great difficulty, the geophysicists Chandrasekharan Manoj and Stefan Maus, from the National Oceanic and Atmospheric Administration Agency (NOAA), in the US, along with Arnaud Chulliat, at the Paris Institute of Earth Physics, in France, managed to visually distinguish a 1 nanotesla signal picked up by a magnetometer on Easter Island, 3,500 kilometers away from the earthquake’s epicenter. The signal coincided with the tsunami’s arrival at the island and its intensity, according to the calculation published by the researchers in the American Geophysical Union’s newsletter EOS, on January 11, 2011, was consistent with the height of the wave on the open sea detected by the underwater pressure sensors (15 centimeters).
The article caught the eye of Virgínia, who, with supervision from Mendes and the geophysicist Andrés Papa, from the UN, works with the analysis of geomagnetic disturbances resulting from the interaction of the Sun and the Earth, which are registered by the Vassouras Observatory (RJ) and by the International Real Time Magnetic Observatory Network (Intermagnet). Brazil is located in a very peculiar region from the geophysical point of view: it is influenced by the Southern Atlantic Magnetic Anomaly, the Equatorial Electrojet and the Equatorial Ionization Anomaly (or Appleton Anomaly). These phenomena make the effect of the disturbances in the magnetic field more complex over Brazilian territory, and this can get in the way of prospecting for minerals and affect electric energy transmission lines.
The scientists realized that they could use a numerical method developed more than six years earlier to study geomagnetic disturbances in the search for signs of this type associated with tsunamis. The mathematical technique is known as “wavelet analysis.” It is widely used by physicists and engineers to distinguish localized structures or, stated more colloquially, “needles in haystacks.”The tool acts like a microscope that is capable of zooming in on the characteristics of signs that would otherwise go undetected. This property makes it possible to identify local irregularities in the geomagnetic signal, including among others the start of a tsunami and the typical signature of its spread.
Utilizing this technique, Virgínia, Mendes and Papa, together with Inpe’s wavelet specialist, Margarete Domingues, analyzed the data from the stations in the Indian and Pacific Oceans that are part of the Intermagnet network, which is maintained by 44 countries, including Brazil, and which makes its data available over the internet. For three recent tsunamis (the Japanese one of 2011, the Chilean one of 2010 and Sumatra-Andaman one, which on December 26 of 2004 caused the death of almost 300 thousand people in various countries that border the Indian Ocean) the researchers found magnetic signals preceding the arrival of giant waves at 10 of Intermagnet’s stations.
Virginia recalls that it was not easy to find magnetic stations close to the tsunamis’ points of origin, especially for the events of 2004, which hit a region of poor countries, where there are few stations, and that of 2011, which occurred so close to the coast that there was interruption in the supply of data from the closest observatory, at Kakioka, in Japan. The fact that they were not always able to obtain data from coastal stations equipped both with magnetometers and with marigraphs also made it harder to produce a more detailed comparison between magnetic signals and the level of the sea. The exception was Papeete station, in French Polynesia, which is equipped with both instruments. At that station it was possible to pick up magnetic signals of the 2010 Chilean tsunami up to two hours before the wave arrived.
How a tsunami is born
Usually produced by abrupt shifts in fault lines on the seabed (which also cause earthquakes), tsunamis begin as waves hundreds of kilometers long. Starting off in deep waters, they spread quickly, crossing the oceans at speeds of 600 to 800 kilometers an hour, but only rise a few dozen centimeters above sea level, often going unnoticed even by boats and ships. However, when they reach the coast, the change in depth results in a radical transformation in their format; the length of the wave decreases, its speed drops and, what is most impressive, its height grows; it is possible for the tide to reach dozens of meters in height.
Since not every oceanic earthquake causes a tsunami, the seismographs that are spread around the globe are insufficient to warn populations in risk areas. This is why there are dozens of pressure sensors installed on the sea bed, most of which are in the Pacific Ocean. However, only the richest countries have enough resources to pay for the installation and maintenance of the sensors, which means that various coastal populations are vulnerable. Furthermore, the system can take hours to identify a tsunami and does not always manage to produce an accurate calculation of its size. For example, on March 11 of this year, a Japanese weather bulletin warned of the arrival of a tsunami with waves at least 3 meters high, whereas in fact the waves went as high as 50 meters.
It may be possible to overcome some of the limitations of the current tsunami warning system by adopting the approach advocated by the Brazilian scientists. The geophysicist Maurício Bologna, from the University of São Paulo, who did not take part in the work of the Inpe-UN team, notes that magnetic sensing has “a significant advantage” over underwater pressure sensors: the capacity to determine not just the size of the waves, but also their direction, which would help to calculate the properties of the tsunamis in real time. Bologna also highlights the low cost of the method, which would make use of Intermagnet’s existing observatories. Moreover, the construction of new stations on land would be a lot cheaper than the installation of sensors on the seabed.
For the geophysicist Robert Tyler, from the US Space Agency Nasa, the work that the Brazilian scientists have carried out is both “important and timely.”Tyler explains that the method developed could be used to analyze data from the European Space Agency’s Swarm Mission, which in 2012 will launch three satellites designed to measure geomagnetic variations caused by alterations in the ocean currents. “The flows of the oceans play a central role in the changes of the climate system as well as in natural disasters, such as tsunamis,”he explains.
Analysis of the characteristics of the electrodynamic solar plasma-magnetosphere coupling based on the effects on the planetary electrical currents (nº 2007/07723-7); Modality Regular Support for Research Project; Coordinator Odim Mendes Junior – Inpe; Investment R$ 44,274.95 (FAPESP)