[ad_1]
Popular mechanics
When preparing for a natural disaster, every second counts. Any announcement of an impending disaster leaves one more second to residents to prepare for and prepare for the future. Some natural disasters, such as earthquakes and tsunamis, are actually equipped with warning systems: localized changes in geomagnetic fields. Scientists at the Tokyo Metropolitan University are developing a deep neural network to understand the changes in magnetic fields.
The magnetic field of an earthquake is created by phenomena called piezo-magnetic effect, or piezomagnetism. Earthquakes mitigate the stresses accumulated between tectonic plates. Thus, the symmetry of the elements of the Earth's structure can become out of sync. When they become asymmetrical, the geomagnetic field around the area changes.
With tsunamis, their rapid movement across the ocean causes changes in atmospheric pressure. When the atmospheric pressure begins to alter the ionosphere, the geomagnetic field changes at the same time.
These are two very different processes, but they have the same result. One or the other can be detected by the observation and it is recorded faster than practically any other activity of the planet: electromagnetic waves move at the speed of light. It is possible to understand almost instantaneously any change in the geomagnetic field of the Earth, anywhere in the world.
However, the Earth is a living planet. This means all kinds of minimal changes and changes throughout the day, leaving scientists with the classic problem of trying to find the signal through large areas of noise.
Professors Yuta Katori and Kan Okubo, who are doing measurements around Japan, have made estimates of what a normal geomagnetic field looks like in the region. To give an idea of the amount of geomagnetic flux existing in the world: in their Japanese observation sites, in 2015, Katori and Okubo were able to accumulate 500,000 data points.
They then began to feed these data points into a neural network. With their data points, they can now estimate what a normal magnetic field looks like in the region, providing a better understanding of what an abnormal reading due to an earthquake or tsunami would look like.
The hope is to associate the neural network with high sensitivity detectors, which allows a rapid emergency alert. Given that failing emergency alert systems have worsened the recent tsunami in Indonesia, the sooner the neural network becomes operational, the better.
Source: Tokyo Metropolitan University
<p class = "canvas-atom canvas-text Mb (1.0em) Mb (0) – sm Mt (0.8em) – sm" type = "text" content = "(# You might also like,,)"data-reactid =" 32 ">(# You might also like,,)
Source link
