The Valencian Research Institute for Artificial Intelligence (VRAIN), the Institute of Water and Environmental Engineering (IIAMA) of the UPV, and the Valencian Graduate School and Research Network of Artificial Intelligence (ValgrAI) have developed a hydrological digital twin capable of anticipating potential barranco overflows. The tool uses real-time rainfall data and allows tracking the predicted evolution for the next 48 hours.
During the presentation, a simulation of the PGTEC tool's application in the Poyo barranco basin was conducted, using rainfall data from the storm on October 29, 2024. The PGTEC system (Platform for the Management of Early Prevention of Climate Emergencies) integrates hydrometeorological data from multiple official organizations, unifying them in a common, interoperable space for intelligent management.
A direct application is the development of digital twins that predict barranco water levels and represent on a map how the water would advance, providing key information to gain time in an emergency. PGTEC incorporates digital twins for the basins of L'Horta Sud and the Carraixet barranco.
The system combines the TETIS hydrological model, which estimates water flow and potential overflows based on rainfall data and weather forecasts, with HIDRIS, which reproduces how water would spread across the terrain, its depth and speed, and identifies critical infrastructure affected.
Instead of interpreting numerous separate maps and forecasts, emergency managers can consult the information in a visual environment. TETIS highlights the most compromised sections with colors, allows tracking the predicted evolution for the next 48 hours, and runs each simulation in approximately one to two minutes. It also allows simulating historical episodes or altering storm parameters.
Additionally, a simulation of the October 2024 DANA storm in Valencia was presented, transferred to the Carraixet barranco to assess its impact on neighboring populations.
PGTEC also includes WATER4CAST, a tool that applies an artificial intelligence algorithm to improve weather forecasts and help anticipate droughts and conditions favorable to forest fires, addressing emergencies that require planning from a few hours to months in advance.
Félix Francés from IIAMA emphasizes that the project addresses both short-term emergencies like floods and long-term risks like droughts, relying on the same interoperable data space.
Vicent Botti, director of VRAIN and general director of ValGRai, concludes that PGTEC brings together very different organizations in a single environment, allowing them to connect their data for a comprehensive view of the situation and overcoming data silo issues.
The PGTEC project has been funded by the Ministry for Digital Transformation and Public Function, within the framework of actions promoted by the Directorate-General for Data and the Data Economy Reference Centre (CRED).
