The Central Water Commission, in India, has partnered with an Artificial Intelligence system to feed with data a model that can predict floods.
India is one of the countries that suffers most from floods throughout the year. These phenomena are often relegated to certain dates. But beyond a more or less broad temporal range, little is known about when and, above all, where the torrential rains that wreak havoc will take place.
The Indian government was therefore interested in using all the means at its disposal to better manage such situations. For this, the Central Water Commission, which manages water resources at the state level, has relied on artificial intelligence. The organization has started collaborating with Google to feeding algorithmic models that can shed light on potential catastrophes in the future.
For a while now, the model has been tested in the city of Patna north-east of the country, on the banks of the Ganges. But now the technology will expand to the rest of the country. It’s basically about bringing together the knowledge and data of the Central Water Commission with Google’s artificial intelligence technology.
Historical events, water level readings, ground data and area-specific elevation are used to feed the flood prediction algorithm. This information allows experts to create models that estimate more accurately that so far the when and where there will be floods. Moreover, the system will be able to calculate the magnitude of the flood.
Getting ahead of nature
The use of artificial intelligence to predict natural phenomena that can turn into disasters is not new. Several companies specializing in artificial intelligence, such as IBM or Microsoft, have worked in these fields. And so have smaller companies with a more concrete approach.
The truth is that between them-although sometimes the public sector has also been the driver of the initiatives- have tried to predict earthquakes, hurricanes or floods. California has been one of the places where the most effort has been put into earthquake prediction. While in Japan they investigate how to know when torrential rains will occur, which can cause great damage to the country.
In India for the moment they have focused on floods, which every year they cause fatalities and significant material damage. If the system is successful it could be exportable to other places.
If you are interested in AI do not miss how Aura, the AI of Movistar Home works.
Images: Oxfam International, EU Civil Protection and Humanitarian Aid Operation