Google announces 1st Flood Warning System based on AI and Machine Learning in Brazil

O Google, in partnership with the Geological Survey of Brazil (SGB-CPRM), announced the first Flood Alert System based on Artificial Intelligence and Machine Learning to operate in Brazil. The novelty joins the other solutions of the Google warning against environmental disasters, such as the Alerts
SOS and Public Notices.

Unveiled during the Understanding Risk Forum, organized by the World Bank, in Florianópolis (SC), the new technology, which is already available in other countries, such as India and Bangladesh, will now also be available in Colombia, Sri Lanka and also in 15 African countries , including Chad, Nigeria, Republic of Congo and South Africa.

How Google's New Flood Prediction System Works

The system created by Google uses the combination of several metrics, such as water levels in rivers, meteorological indicators and satellite images, in addition to using computational power and experience in machine learning to develop automated flood forecasting and warning systems on a global scale.

In Brazil, the system will be implemented in two stages, in partnership with SGB-CPRM, an agency linked to the Ministry of Mines and Energy (MME)🇧🇷 The first will issue real-time riverside flood alerts to more than 60 locations across Brazil – showing where floods are occurring in real time. The second step takes place in the use of flood forecasts generated by the SGB-CPRM hydrology team and by the systems of the Google in selected areas of Brazil.

Playing a key role in the new system of Google, Geological Survey of Brazil (SGB) currently operates Hydrological Alert Systems in 17 hydrographic basins throughout the Brazilian territory, with automatic monitoring of rainfall and river levels, maps of hydrological risks, and sending of monitoring and alert bulletins with flood forecasts. Information from this monitoring is periodically sent to civil defense agencies, state and municipal public agents, in addition to
of the general population.

rains in Brazil

Recognized as an essential service to society, the Alert Systems of the SGB-CPRM now join the technologies of Google at a propitious time. This is because, in recent years, Brazil has seen a growing number of natural disasters. Earlier this year, for example, floods hit the state of Bahia, leaving more than 30 people homeless and 70 homeless. The total number of people affected exceeded 800.

Still in the same period of the year, the state of Minas Gerais also suffered from heavy rains. At the time, the Civil Defense of the state disclosed that more than 400 municipalities were in an emergency situation, while accumulating more than 20 deaths, 8133 homeless and 48.930 displaced due to the rainy season.

According to a United Nations report, Brazil is one of the countries with the population most exposed to the risk of riverside flooding in the world. Between 2000 and 2019, more than 70 million people were affected by floods in the country. In this sense, solutions such as the flood forecasting system can help local governments, entities and the population to prevent natural disasters.

The solution has already proved to be functional and of great help in several countries. In India and Bangladesh, for example, the Google sent 115 million flood notifications and alerts to 23 million people during the past year.

See also:

Innovators Hive: Google talks diversity and inclusion at developer event.

Source: Google.