Experts agree that global warming is the main cause of more frequent and intense flooding. According to data published by Oxfam Intermón, the number of disasters caused by floods and droughts in the ten most affected countries has increased dramatically, from 24 in 2013 to 656 in just ten years. In this context, the key to minimizing flood impact lies in proper risk management encompassing the essential phases of mitigation, preparedness, response and recovery.
October 2024 will go down in Spanish history for the devastating effects of unprecedented flooding that affected almost 80 Spanish municipalities. According to the Spanish Home Office, convective rainfall, such as flash flooding, is caused by increased heat on the Earth’s surface and has the potential to cause flooding in rivers and urban areas, with devastating consequences for both infrastructure and human life.
Therefore, in situations where speed and precision are essential, “the preparation and response phase must be carried out within strict timeframes to enable crucial decisions to be made”, highlights Sergio Morant, Floodrisk Specialist at Xylem Vue. This is where “Early Warning Systems (EWS) become invaluable allies, operating with maximum precision and efficiency, improving predictive capacity and ensuring effective decision-making, protocol activation and automated responses”.
Forecasting, modelling, automation and coordination
The activation of emergency protocols and plans is based on accurate hydrometeorological information. This involves combining different technologies and sensors that enable real-time monitoring and response. Advanced weather forecasting systems, such as the HARMONIE (AEMET), IFS (ECMWF) and GFS (NOAA) numerical models, predict extreme weather events between two and fifteen days in advance, while nowcasting, based on the use of radar technology and short-term forecasting algorithms, offers high-resolution forecasts in real time and up to three hours ahead. These tools facilitate the early detection of convective phenomena, which is crucial for river and urban flooding management.
Another aspect to factor in when anticipating and managing floods is the use of hydrological and hydraulic models for risk assessment, as Morant outlines. Hydrological models transform rainfall into flow, simulating the impact of runoff in rivers and wastewater networks. In turn, hydraulic models simulate the movement of flows in riverbeds and sanitation networks, interacting with the digital terrain model (DTM), which helps identify flood-prone areas.
Integration with GIS systems provides greater accuracy in the identification of risk areas, where hyperscale plays a crucial role, as it supports scenario modeling from small urban basins to large hydraulic systems, optimizing decision-making and resource allocation according to the magnitude of the event.
In this context, “rapid response is essential to minimize damage” points out the specialist. Early Warning Systems therefore not only predict floods, but also automatically activate warning systems that restrict access, initiate evacuations and protect key infrastructure. In the opinion of the expert from Xylem Vue, “without automation, response capacity to rapidly evolving meteorological phenomena would be severely limited”. These systems are based on thresholds for precipitation, flow rates and water levels, which are established by experts in the different management units, such as flash-flood basins, reservoir watersheds and basins classified by time of concentration, and are fed by data and predictive models in real time.
Alert levels can be established by intelligently grouping these warnings, linking them to specific emergency plans. Authorities can activate or deactivate the corresponding protocols, depending on the severity and extent of the event, ensuring a coordinated and efficient response.
Implementing EWS enables flood maps to be generated, specific actions to be recommended for key infrastructures such as dams and storm tanks, and personalized alerts to be issued to the relevant authorities. These systems provide advance information on the magnitude of the event, facilitating more efficient emergency management and preventing response breakdowns. Furthermore, an integrated approach to these technologies improves coordination between organizations and reinforces the resilience of critical infrastructures. “Investing in EWS is essential, especially in regions with a history of flooding and in those affected by climate change, as the benefits in terms of safety and protection far outweigh implementation costs, making them key tools for modern flood risk management” highlights Sergio Morant.
In Spain, organizations such as the Ebro Water Authority, the Government of Navarre, and the utility Aguas de Calpe have already deployed these types of digital systems and platforms.
Solutions such as Xylem Vue (XV), developed as a result of the partnership between Idrica and Xylem, already feature hydrometeorological sensors and the results of hydrological and hydraulic models in river and urban areas, including the simulation of reservoir inflows, to trigger warnings and alarms in real time. The solution incorporates specific modules and algorithms that identify and manage alerts for extreme events in real time.
These types of EWS combine detailed information on topography and terrain characteristics, cadastral and demographic data, together with advanced hydrological models, to represent the hydrological-hydraulic behavior of basins and drainage networks in the event of heavy rainfall. This means they can anticipate which areas are prone to flooding and the severity of the possible impacts. According to Morant, they enable monitoring and also “transform data into action, which is key in flood management”,
The ebro water authority and aguas de calpe, examples of anticipation
The implementation of the Xylem Vue platform’s Early Warning System (EWS) solution in the Ebro Water Authority (CHE) has provided an early flood warning system for a range of scenarios.
The application uses cloud computing and big data analytics for real-time, deterministic, and probabilistic hydrometeorological prediction models. It also applies advanced geo-statistics (kriging) and machine-learning techniques (Kalman filtering algorithms) to improve data quality and uses its AI inference engine and heuristics to generate management recommendations. The system runs every ten minutes for advanced preemption.
In addition, the application known as “VIGILAEbro” at the CHE uses preliminary calculations, such as Areas at Risk of Significant Flooding (ARPSis), associated with different return periods to quickly identify areas that may be affected by current and predicted hydrometeorological conditions. This significantly reduces computational processing times, enabling immediate recommendations for dam operations to minimize risks and damage. It also helps to disseminate warnings and alerts to mobile phones and websites according to the type of users and their location. The application provides early, automatic, real-time information on potentially affected infrastructures and facilities.
Another example is the implementation of the Xylem Vue Real-Time What-If Scenarios digital twin in the Aguas de Calpe network to predict urban flooding events, identifying the areas affected, the levels of flooding and how they evolve over time. The system also helps to detect and predict spillways, simulate rainfall scenarios (what-if) with real-time network information, and provide a decision support system. These solutions have enabled the company to detect 26 early warnings of urban flooding in one year.
