Pilot Program Pushes Boundaries for Wildfire Detection

The most recent wildfires throughout western North America have re-emphasized the need for early wildfire detection and validation. Time and resources are in short supply for emergency responders, who have a short window to detect and assess wildfires to help prevent the most damage.

“Time and resources are both in short supply during the critical first operational period,” says  Chief Dave Winnacker of the Moraga Orinda Fire Department. “This period, before large incident management teams have been mobilized, is both when the greatest damage occurs and when the least information is available to guide evacuation and firefighting deployment decisions.”

To help fire departments respond during this critical period, Planet, UC Berkeley (UCB) and the Moraga Orinda Fire District (MOFD), with support from the Gordon and Betty Moore Foundation, launched a pilot to determine if a multi-system approach could be beneficial for early wildfire detection. The pilot’s objective was to use on-the-ground fire sensors to detect a fire and then automatically task a satellite to capture high resolution imagery of the burn location.

They discovered that on-the-ground sensors could automatically trigger the collection of satellite imagery, which can be used by incident management teams to better understand the extent of a wildfire.

Over several trials, the MOFD ignited prescribed fires in oak woodlands where the sensors had been installed. These sensors were made to detect an increase in temperature and a drop in relative humidity, and to send a signal to an interface called BurnMonitor, alerting responders to a potential fire.

A Planet Skysat image captured on May 1, 2020 in response to the tasking request triggered by the motes. The image is overlaid with the positions of the motes in the testing area, with the motes that detected the fire highlighted. © 2020, Planet Labs Inc. All Rights Reserved.

The team at UC Berkeley’s Department of Electrical Engineering and Computer Sciences developed a script that syncs the signals from the sensors with Planet’s SkySat Tasking API. The Tasking API is a programmatic interface that enables users to manage and request imagery collection in an efficient and automated way. High resolution SkySat imagery was captured and delivered back to the MOFD for review.

“[The pilot] is a major milestone towards the development and fielding of these tools,” Chief Winnacker says, “and is indicative of the great potential for improvements that lie ahead.”

To learn more about the compelling results of the pilot, check out our case study. Also, dive into this Fast Company piece covering how valuable this tech could be for detecting fires as soon as they start.