American tech giant Google has stepped forward with its initiative to utilize AI in forecasting floods on a global scale. The company published a research paper in the scientific journal Nature, highlighting AI’s potential in saving lives and limiting damages in flood-affected areas. The AI models have been developed by the team at Google Research.
According to the paper, using AI-based hydrologic technologies can drastically improve flood forecasting even in areas where there is limited flood-related data. “We found that AI helped us to provide more accurate information on riverine floods up to 7 days in advance. This allowed us to provide flood forecasting in 80 countries in areas where 460 million people live”, the paper claimed.
See Related: Bank of England’s Journey Towards Better Economic Foresight
AI-based Hydrologic Technology
The hydrologic model has been trained using publicly available data such as soil attributes, streamflow gauges, and weather forecasts. It uses two Long Short Term Memory (LSTM) networks – a hindcast unit and a forecast unit. The hindcast unit analyzes geophysical data from over a year in the past and sends it to the forecast unit. The forecast LSTM then combines this data with the weather forecast for the next seven days to make highly accurate streamflow predictions.
“Our goal is to continue using our research capabilities and technology to further increase our coverage, as well as forecast other types of flood-related events and disasters, including flash floods and urban floods”, Google stated.
As of 2024, Google’s hydrologic model covers 80 regions across Africa, Asia, Europe, and both South and Central America. The relevant data are available on the Flood Hub platform.
