Imagine waking up to news of a massive earthquake shaking the city of Mandalay, Myanmar. It’s a daunting scenario, but thanks to cutting-edge technology, we’re better equipped to respond quickly and effectively. In the early hours following this catastrophic event, advanced satellite technology was deployed to assess the damage. A powerful 7.7 magnitude quake had hit near the city, leaving devastation in its wake. The mission? To capture images from space and use artificial intelligence to pinpoint the hardest-hit areas needing immediate assistance.
One of the hurdles, as explained by Juan Lavista Ferres, Microsoft’s chief data scientist, was the cloud cover. “There’s no way to see through clouds with this technology,” he shared. Despite this challenge, the team was determined. By late Friday evening in Redmond, Washington, satellite images from Planet Labs were being analyzed by Microsoft’s AI for Good Lab. This wasn’t their first rodeo—they’ve tackled disasters like Libya’s floods in 2023 and the wildfires in Los Angeles. But this time, they crafted a custom AI model specifically for Mandalay’s situation. “The Earth is too different, the natural disasters are too different, and the imagery we get from satellites is just too different to work in every situation,” Lavista Ferres noted.
The analysis was eye-opening. It revealed that 515 buildings had suffered 80-100% damage, while 1,524 buildings experienced 20-80% damage. This crucial data helps direct rescue teams to where they’re needed most, though Microsoft stresses the importance of verifying these insights on the ground. These findings have been shared with humanitarian organizations like the Red Cross to bolster their relief efforts.
Planet Labs, with its fleet of 15 satellites, has been keeping a watchful eye on the affected regions in Myanmar and neighboring Thailand since the quake hit. This continuous monitoring is vital for ongoing relief operations. It’s a testament to how technology, when used thoughtfully, can make a real difference in times of crisis.
