We are pleased to announce the launch of our project titled “Spatio-temporal modeling of groundwater recharge and low-flow conditions under climate change using a physics-informed graph neural network.”
This project aims to develop and operationalize a physics-informed artificial intelligence framework to simulate groundwater recharge and low-flow conditions in southern Québec under changing climate conditions. A key objective is to better represent surface–groundwater interactions and daily hydrological processes (e.g., snow dynamics and freeze–thaw cycles), and to explicitly quantify and propagate uncertainty across climate scenarios.
The project will directly support the Southern Québec Hydroclimatic Atlas by providing robust, interpretable, and regionally scalable indicators to inform water resources planning and climate adaptation strategies.
This project is part of the QClim’Eau initiative, a collaboration between the Ministry of the Environment, the Fight Against Climate Change, Wildlife and Parks (MELCCFP) and Ouranos.