Fredericton siblings win science prize for developing model to predict flooding

New Brunswick siblings have won a major award for developing a system to predict flooding.

Leonardo Cui, a Grade 11 student at Fredericton High School, and his sister, Cynthia, who's in Grade 10, won the Intact Climate Change Resilience Award at the Canada-Wide Science Fair in Fredericton.

"It's a very useful tool for the public, especially those who live next to the river who are directly affected by flooding," Leonardo Cui said.

"If they can know their house is [going] to be flooded, they can evacuate earlier, they can get their stuff out earlier or they can prepare better with higher barriers."

Leonardo and Cynthia were among 465 students who presented their ideas at the Canada-Wide Science Fair, which ended Friday.

How their model works

The Cuis collected historical climate data from the Canadian government website and used a machine-learning algorithm to forecast flood patterns.

Machine-learning is a method of data analysis that relies on patterns and inference to perform a specific task, like mapping flood patterns. 

The siblings also created a flood map.

"Suppose you were living on a certain street," Cui said. "You could see if it's being flooded or not on our model and also you can also see the severity of this flood."

The judges were surprised by the accuracy of their model.

"We compared our model with the actual water level and the water level the predictions that [the Emergency Measures Organization] made and we found that we were more accurate than EMO 74 per cent of the time," Cui said.

The award means they'll receive $1,000 and an all-expenses paid trip to Waterloo, Ont., to visit the Intact Centre on Climate Adaptation.

"I'd like to see our model be implemented on the field in New Brunswick, maybe all of Eastern Canada," Cui said.

Youth from across Canada flocked to Fredericton for the national science fair, where around $1 million in scholarships and prize money is awarded.