Published 1 day ago • loading... • Updated 1 day ago

Finding Canada’s secret tornadoes

The Northern Tornadoes Project is using new AI to detect tornadoes in remote Canadian forests by scanning satellite imagery for telltale patterns of tree damage that would otherwise go unrecorded.
Despite a $180 million investment in Doppler radar, Environment and Climate Change Canada faces ongoing technical challenges, and warnings were issued for only 20 per cent of tornadoes last year.
Last year, the AI tool confirmed more than two dozen tornadoes in remote areas, including four near White River, Ontario. "A perfect job for AI to do," says David Sills, director of the Northern Tornadoes Project.
Western University wind engineer Greg Kopp advocates for stronger building standards, suggesting that using six-inch screws to secure roof trusses to walls could prevent structural collapse during tornado strikes.
Kopp hopes to incorporate resilient construction guidelines into national building codes by 2030, while Sills emphasizes improved data collection, saying "We can do better" to protect Canadian communities.

Insights by Ground AI