A new study published in the journal Hydrology offers a machine learning model developed by Concordia University researchers to improve flood evacuation protocols.
PhD student Mohamed Almetwally Ahmed and Samuel Li, professor and head of the university's civil engineering department, created a technique that uses artificial intelligence to more accurately predict the short-term flow of a river, which is critical data for the evacuation.
Read more:
Atmospheric river: what is it and which is the biggest in the world? Over Africa, a jellyfish-shaped storm is recorded by satellite. Climate crisis: new technology uses artificial intelligence to predict storms Examines river flow in detail and allows for more effective predictions of possible floods - Image: Hydrology How the study model works The model was developed based on historical data from hydrographic stations and new parameters climatic conditions, such as precipitation, temperature and humidity. The researchers focused on measuring convection, which represents the speed of water movement between two stations on the Ottawa River. The research used decades of data collected by the Canadian government and tested the model with data from other regions, such as the Boise and Missouri rivers in the USA. The model provides accurate estimates of daily flow and especially real-time flow, helping to predict water flow for up to 24 hours, which is essential for effective evacuation. The method, which uses nine predictors (seven climatological and two historical), is adjusted according to the forecast time. Over time, this model should be operational and accessible to the public, providing real-time water level predictions, similar to weather forecasts.
Ahmed's idea is for authorities to use the model as a tool to plan evacuations, optimize transport logistics and save lives and property during floods.
The research could help authorities begin evacuating civilians in time to ensure everyone's safety - Image: humphery/Shutterstock