Inspiration
Fire records don't capture every fire. Small, remote, or short-lived fires can go unreported, creating blind spots in our understanding of environmental damage and carbon emissions. We wanted to use open satellite data to uncover these hidden events.
What it does
Ada's Eye compares NASA FIRMS satellite fire detections with independently reported fire records to identify candidate under-reported fires. It uses statistical anomaly detection and spatial-temporal clustering to distinguish unusual fire-like events, then turns the results into a gamified globe where users can test their ability to spot anomalies.
Challenges we ran into
Finding datasets and figuring out how to find anomalies.
Log in or sign up for Devpost to join the conversation.