Before sending any crews to the scene, the agency checked high-powered cameras on a computer to visualize the plume. The threat was credible, and because high-value crews were able to address the fire immediately, they held the blaze to 23 acres.
The pilot program has been up and running since March of 2023. Twenty-one mountaintop cameras have been installed across the state, deployed in connection with satellite feeds and other sources of fire detection.
“We were the first to alert DNR dispatch,” said Sonia Kastner, CEO and founder of Pano AI. “They later did get a 911 call, but their assessment is that our alert sped up the DNR response time by 20 to 30 minutes because we provided the notification that there was a fire. We also gave them visual information that allowed them to know where to go and gave them situational awareness of the severity of the fire.”
What later became known as the Jackson Road Fire was contained in less than a week, resulting in no loss of life or structures.
That’s a critical feat in the state of Washington, where peak wildfire season is typically sometime between mid-July and mid-September, when the state is experiencing its hottest and driest conditions. In recent years, drought, extreme heat, and high winds have extended the season, putting a strain on crews.
DNR is Washington state’s wildfire fighting force, staffed by about 1,300 employees, including 800 permanent DNR employees and 550 seasonal employees, about 120 wildland fire engines, nine helicopters and six single engine air tankers under contract with DNR.
According to DNR data, about 85 percent of wildfires in the state are caused by humans, making it difficult to predict when and where they will happen. But with 21 new cameras equipped with AI technology monitoring for certain types of smoke, and capable of providing optical zoom feeds to assess fire behavior for the dispatch center, DNR is optimistic that cameras and technology will revolutionize their efforts to protect 13 million acres of private, state and tribal-owned forestlands.
The Pano AI pilot cameras have been positioned in places DNR determined to be at greatest risk. DNR Assistant Division Manager of Plans and Information, Wildland Fire Management Division Angie Lane told Government Technology that so far, the pilot’s results have also been impressive in verifying and assessing reports of fire from human beings.
“We’ve had many examples where the Pano AI camera was critical in that detection, making sure that any reported fire was real and actionable,” Lane said. “The cameras have played a critical role in that extra tool in the toolbox for our early detection, because the only way we are going to get ahead of these fires and the amount of fires we have, and try to keep those fires small, is if we detect them early.”
While the Washington DNR budget only allows for 21 cameras, the vision is to grow a network of wildfire detection cameras with local fire departments and private partners, so the state can build a real-time camera response center.
“We’d have a platform where we’d have multiple camera groups or companies that could see all the camera locations,” Lane said. “The idea is that hopefully, we could click on it and at least view what the camera is viewing.”
There’s also potential, she added, for the cameras to be used to monitor prescribed burns.
“Making sure that the fire stays within that area that we’re trying to control when we’re doing a prescribed fire,” said Lane. “So that’s another way of utilizing the cameras that we haven’t completely dealt with yet. That’s on the horizon for the future.”
