Digital technology to tackle 21st century floods on a global scale
Floods, the deadliest natural disasters worldwide, have claimed the lives of approximately 3.2 million people in the past century; that is more than 50% of the victims of natural disasters in that period, according to data from the Water Research and Technological Development Center of the University of Salamanca (Spain).
Since this year began, we have been witnessing alarming flood episodes in different parts of the world, some of them affecting millions of people and leaving behind multimillion-dollar economic losses.
A wave of severe floods continues to affect the globe in 2024
Last January, the Republic of Congo experienced the worst floods in the last sixty years, with an estimated 1.8 million affected people.
A few weeks later, news arrived of the devastating toll that torrential rains in Kenya had taken since mid-March, with 228 dead and hundreds of thousands displaced, as the forecast still doesn't give respite to most Kenyan counties.
The phenomenon of El Niño has not only intensified the rainy season in this African country, but, according to the United Nations Office for the Coordination of Humanitarian Affairs (OCHA), more than 637,000 people have been affected by precipitation in the eastern zone of the continent, including Tanzania, Rwanda, Burundi, and Somalia.
In early April, as a result of rapid thawing, several mighty rivers overflowed through Kurgan and Tyumen in Russia, while several towns located in the Ural Mountains, Siberia, and areas of Kazakhstan near the Ural and Tobol rivers also suffered damage. On April 8, the Russian government declared a state of emergency in three cities in the southwestern region, adding Orsk to Kurgan and Tyumen, after a dam broke following days of torrential rains, an episode that affected more than 4,000 people, including 885 children, who had to be evacuated from the flooded areas of the region.
To these catastrophes, we add the recent floods in southern Brazil, in the state of Rio Grande do Sul, which are already considered the worst natural disaster in the history of the region. The number of deceased people as of yesterday had already risen to 83, while another 110 are missing. The magnitude of this disaster has also prompted the relocation of 115,844 people, 1.2 million households are without power, and another 18,487 people have been accommodated in public shelters.
Likewise, the city of Houston, Texas, has been experiencing a wave of floods since last Friday, which has led authorities to order the mandatory evacuation of the metropolitan area of the city, putting at risk a significant portion of the 7.3 million people residing in this area. These torrential rains have caused the San Jacinto River – flowing between Lake Houston and the Gulf of Mexico – to overflow.
Digitalisation, the greatest ally in flood risk management
Natural disasters of such magnitude and unpredictability, like floods, find in digital technologies the most suitable tools to work on prevention, anticipation, and early warning systems to try to mitigate, as much as possible, their devastating impact. Real-time data collection through sensors and satellite systems allows constant monitoring of water levels and weather conditions, giving authorities the ability to anticipate these episodes. In addition, simulation models and data analysis help assess risks and design more effective management strategies for the future. This synergy between technology and risk management is essential for protecting lives and properties in a world with increasingly frequent extreme weather events.
In the 2023 edition of the Spain Smart Water Summit, Amphos21 presented the benefits of the Aqua Learning software for flood prediction and early warning creation. This tool has a wide range of models of flood-prone areas and their return times, and its algorithm has proven to be able to visualize the flood zone 8,000 times faster than a hydraulic model.
On the other hand, Infoworks ICM by Autodesk Water has shown excellent performance in Türkiye's National Flood Forecast and Early Warning System (TATUS). In a country that suffers about 125 floods per year, being able to predict rains and runoff with up to 72 hours of anticipation will allow early warning and preparation measures.
In addition, Infoworks is being used to manage stormwater overflows. In Spain, recent regulatory changes (update of the Public Hydraulic Domain Regulation) have introduced new requirements for the management of stormwater overflows during and after rainfall events, seeking to mitigate environmental impacts and improve water quality in urban environments. The new requirements include the preparation of Integrated Sanitation System Management Plans, which must include details concerning the monitoring of discharges. InfoWorks ICM can help in this regard, as it enables more efficient and consistent modelling of complex elements in water networks.
In the field of flood risk management, Artificial Intelligence (AI) has transformative potential, thanks to its ability to process large volumes of data and offer more accurate predictions.
Among the AI techniques used in this field, machine learning and neural networks have proven to be especially effective, employing a methodology that allows machines to identify patterns in complex data without explicit programming.
In the context of floods, these techniques allow for a deeper understanding of hydrological and meteorological patterns, essential for effective prediction and management of risk.
A notable example is the Sen1Floods11 project, which uses neural networks to identify flooded areas from radar and satellite images. This project has demonstrated its global applicability by providing a dataset covering flood events on six continents.
Thus, other successful cases in the application of AI in flood management include the use of photographs to estimate flood depth during Hurricane Harvey in Texas, Google's flood forecasting project, and an AI model in Cork, Ireland, to accurately predict floods.
The possibilities offered by digital technologies in the field of prediction suggest a promising future for flood prevention and risk management. The combination of real-time data, advanced predictive models, and early warning systems are crucial tools for authorities and managing entities to anticipate these catastrophic events, a solution that can make a difference in their impact on human and material losses. As we move towards a future where extreme weather events are the new normal, investment in digital technologies and the application of innovative solutions become cornerstones for building more resilient and safer societies against floods and other climate challenges.