On the morning of June 9th, Laura Lin was working remotely from her home in Lanesville, a rural Indiana community located approximately 15 miles from the Kentucky border. Unbeknownst to her, heavy rainfall outside was rapidly transforming her property.
Lin recalls being mid-conversation on a Zoom call when she noticed debris from her woodpile floating near the barn. “I was like, ‘What?’ And I immediately was like, ‘I have to go,'” she said. She shut down her laptop, woke her children, and evacuated the house, sensing something was critically wrong.
The family managed to reach safety at a neighbor’s residence. However, Lanesville subsequently recorded more than 8 inches of rain in just a few hours, vastly exceeding the three-inch threshold typically associated with severe flooding.
Incidents like this highlight the unpredictable nature of flash floods, which can strike with little to no warning. New technology leveraging satellite imagery and machine learning is being developed to change that dynamic by providing earlier detection and alerts.
Eight inches in hours is insane. My cousin lives near Lanesville; hopefully, these systems save lives soon.
Machine learning on satellite data sounds promising, but early detection depends entirely on model accuracy and latency.
Does anyone know if this works in rural areas with poor satellite coverage? That feels like a major gap.
Laura’s story is terrifying. Five minutes could have been the difference. We need this tech now.