Hi-tech cameras orbiting Earth and on the ground, along with AI, help firefighters extinguish blazes before they spread
The fires in Spokane, Washington, this summer are the latest reminder that when a wildfire is detected can be just as important as where it starts. A blaze found within minutes can often be contained before it spreads. Wait too long, and a small spark can become a major disaster.
A new effort to shorten that timeline began in July when a SpaceX rocket carried the first three FireSat satellites into orbit. The trio is the first part of a planned network of 50 satellites built solely to spot wildfires while they are still small enough to stop.
FireSat is designed to fill an important gap. Existing weather and Earth-observing satellites can detect large fires, but they often lack the resolution or revisit the same location too infrequently to spot small fires shortly after they ignite. FireSat is designed to detect fires as small as a beach bonfire and, once the full constellation is deployed, scan every point on Earth about every 20 minutes.
The satellites use infrared sensors and artificial intelligence to search for heat that could signal a new wildfire. The system compares new images with earlier ones and accounts for factors such as weather conditions and nearby sources of heat before sending alerts to emergency responders. The goal is to identify real wildfires while reducing false alarms.
FireSat is designed to work alongside a growing network of cameras on the ground that use artificial intelligence to search for smoke and heat. Together, the systems could give firefighters their earliest warning yet that a fire has started.
Wildfires have become one of the most destructive natural disasters in the United States. Rising temperatures, prolonged drought and more frequent stretches of hot, dry and windy weather have made fires easier to ignite and harder to contain.
Climate Central reports that some parts of the western United States now experience about two more months of fire weather each year than they did in the 1970s.
High in the mountains west of Denver, two technicians climbed a 150ft cell tower to inspect a pair of wildfire-detection cameras built by Pano AI.
“There are two Pano AI cameras,” said Arvind Satyam, co-founder of the San Francisco startup. “They’re sitting on top of that cell tower.”
Pano AI has installed more than 1,400 cameras across 17 states. The cameras continuously scan the landscape from towers and mountain tops. Artificial intelligence reviews every image, searching for smoke during the day and heat after dark.
“The AI algorithms are being run continuously on that imagery to say ‘smoke, not smoke,’” Satyam said. “At night-time, we’re looking at heat signatures, so heat versus cold. And then we’re able to zoom in and validate that it is a potential fire start.”
If the system spots a possible fire, a human analyst reviews the images before an alert is sent to fire agencies. That process often takes only a few minutes.
Firefighters say that head start can make all the difference.
“It’s a great tool to validate what’s going on,” said Brendan Finnegan, assistant chief with West Metro fire rescue near Denver. “Having those eyes that can alert us well in advance of potentially a 911 call make it a valuable tool for us.”
