At a time when natural disasters are increasingly being reported, counteracting them is a major topic of discussion at various levels. On today’s International Day for Disaster Reduction, let’s look at how Everday AI can help detecting disasters and also save lives and minimize potential damage.
Everyday AI: Detecting Disasters
Early Earthquake Detection
Severe earthquakes and tsumanis need to be detected as early as possible – that’s clear. Until now, this has been done by evaluating seismic waves, which are recorded and analyzed. However, this method is inaccurate – with far-reaching consequences, as the great Tōhoku earthquake in 2011 taught us. That earthquake was classified at level 8 instead of level 9, which meant, among other things, that the resulting tsumani was 15 meters high instead of 3 meters (source: ZAMG, 2021).
Geophysicists are now increasingly concerned with gravitational signals: if the Earth’s masses shift, fluctuations in gravity occur. This produces also waves, known as elastogravitational signals, which can be measured at the speed of light and thus faster than seismic waves. This is precisely where researchers come in. Andrea Licciardi and his team rely on AI here, because it can filter out and process even the weakest signals. To this end, Licciardi trained the AI with existing data from past earthquakes in an earthquake-rich region of Japan. Additional data was created. The results speak for themselves. This method was also able to correctly determine the magnitude of the Tōhoku earthquake, among other things. The disadvantage: the AI currently has to be trained individually for each region to be investigated
Andrea Liccardi and his team are currently working with data from other locations in order to be able to test this method – successfully – at other sites and then use it. By working with elastogravitational signals and the faster evaluation, valuable minutes can be gained for. (More about this: Welt der Physik, 2022)
If an earthquake has occurred and, for example, a city has been affected, artificial intelligence can also be used in a different way. Using satellite images, existing and current images of the city can be superimposed. This results in a new city map, which shows emergency forces and helpers, but also evacuees, currently free streets (more on this: CGI, 2021).
But it is not only early earthquake detection that supports artificial intelligence. AI can also be of assistance here:
AI is fed and trained with data from various subsectors on precipitation and runoff of water from rivers. It can help emergency responders move out faster and minimize damage. Read more here: https://www.derstandard.at/story/2000111843674/wissen-wann-das-hochwasser-droht
Using AI, farms can conserve water while still providing optimal care for their crops. Read more here: https://www.forum-csr.net/News/14718/MitKIinderLandwirtschaftWassersparen.html
Detecting forest fires
Artificial intelligence detects smoke through satellite imagery. It even distinguishes whether these are intentional fires, such as campfires, or unintentional, actual fires. As a result fire departments and other emergency services can intervene more quickly. But preventive action can also be taken if artificial intelligence can classify the likelihood of a forest fire from tree population, rainfall and other factors. Smoke cloud detection from space (satellites): https://austria-in-space.at/de/news/2021/ki-rauchwolken-all.php Successful smoke detection in Brazil: https://the-decoder.de/ki-klimawandel-waldbrand-frueherkennung-kann-katastrophen-verhindern/
Advance warning of landslides
Through a variety of images and measurements, Artificial Intelligence can also detect movements of the earth and therefore warn in time of hand slides. Read more here: https://www.it-boltwise.de/erdrutsche-besser-vorhersagen.html
Without a doubt we can conclude that everyday AI will play an increasingly important role in detecting disaster and prevention. It just needs people with a vision and people with the right know-how. Also there will never be a lack of fresh ideas thanks to ingenuity and progress.