A new interactive tool that uses satellites and artificial intelligence creates more accurate wildfire maps for the public and firefighters.

A new interactive tool that uses satellites and artificial intelligence creates more accurate wildfire maps for the public and firefighters.

Fire near Big Fall Creek Road in Oregon in 2017 (Photo by Marcus Kauffman on Unsplash)

Anxiety-provoking images of misty skies from wildfires and burning hillsides appear more and more on my Twitter and Instagram accounts. The photos are posted by friends and colleagues concerned about their homes, loved ones, and thwarted vacation dreams.

As hundreds of wildfires have raged across the western US this summer, a new interactive mapping tool is available for firefighters and residents to track and respond to fires.

The RADRFIRE The tool uses infrared satellite imagery and artificial intelligence to create detailed wildfire maps to track and forecast fires. It was developed in the Pacific Northwest National Laboratory (PNNL) in eastern Washington, in consultation with numerous agencies responsible for fighting fires, a job that becomes increasingly difficult with worsening droughts and climate change. The pirate fire currently burning in southern Oregon is so fierce that generating its own climate.

“We are seeing much bigger fires these days. They are much more dynamic, they move faster. We cannot always use the traditional methods that we have used, ”said the RADRFIRE project leader. Andre coleman. “We need to improve the game a bit.”

Drawing on their experience in other areas of natural disaster response, PNNL’s 14-member team began actively developing the tool less than a year ago, in September 2020. It was launched in May and updates will arrive in August.

RADRFIRE received information from the US Forest Service, the US Bureau of Land Management, the Washington Department of Natural Resources, and Cal Fire.

A new interactive tool that uses satellites and artificial intelligence creates more accurate wildfire maps for the public and firefighters.

Image of the Bootleg Fire in Oregon generated with a new mapping tool from the Pacific Northwest National Laboratory. (PNNL image)

Here are some details of the project, as recently described to GeekWire by Coleman:

Why build this tool? In the past, to track active fires, an incident commander would request that an aircraft fly over a site after dark to collect imagery, an exercise that is limited by aircraft availability, safety concerns, and the ability to penetrate the deck of smoke. . The images were passed to an analyst to draw the maps by hand so they were ready for use in the morning.

Firefighters have used satellite imagery before, but the use of infrared images and higher resolution sensors makes it much more powerful. (In previous satellite images, a pixel could span 2 kilometers; now that’s up to 30 meters.)

One of the capabilities of the new tool is to identify small point fires that are caused by lightning before they become larger and more difficult to fight.

A new interactive tool that uses satellites and artificial intelligence creates more accurate wildfire maps for the public and firefighters.

Andre Coleman, PNNL’s RADRFIRE project leader. (PNNL photo)

What is the information used for? The images help commanders make decisions about the best way to deploy their resources, keep firefighters safe, predict the spread of a fire, and identify structures in their path. They can help municipalities and others calling for evacuations of people in distress.

It also helps utilities understand what infrastructure is at risk. Power companies have limited access to areas with active fires, so the images can keep them better informed and help with quicker responses once they regain access. The Bootleg Fire, for example, has threatened essential power transmission lines that bring electricity to California.

What about the headlines grabbing fires? Because it is automated, the RADRFIRE system can extract and analyze information on small and large fires alike. Fire suppression resources are concentrated in the largest and most complex type 1 fires, but there are type 2, 3 and 4 fires that also affect communities and can be more remote. Making smart use of the few resources available for smaller fires is crucial.

“This is really a huge benefit to the people who work on these Type 3 and Type 4 fires,” Coleman said. “They get a lot less attention, a lot fewer resources to work on them.”

Whose satellites are they? Who owns the system? The tool uses information from national and international government satellites with open access. The team was concerned that using commercial satellites would become too expensive and limit who could use the tool.

“We just feel that this should be open access information,” Coleman said. “Someone shouldn’t have to buy it.”

Wildfire Forecasting: How to Stay Safe and What to Expect This Summer in the Pacific Northwest

Because the team is taking advantage of the satellites of others, they take the images they can get so that some areas are photographed more or less than others and at random times. The goal of the project is to capture a couple of images a day, which are added to the information gathered through other means to create a more complete image.

Coleman hopes the program will continue to be administered by a government agency. The US Department of Energy Office of Artificial Intelligence Technology has supported him in collaboration with the Joint Center for Artificial Intelligence of the US Department of Defense

How does RADRFIRE contribute to the science of firefighting? Working with the Forest Service, the team has developed an algorithm to better plan and monitor the use of fire retardants that are dropped by aircraft over flames. The tool can quickly determine where previous drops were planned, where they actually landed and spread, and their effectiveness. In the past, commanders were able to collect some of this information, but it took more time and resources.

A new interactive tool that uses satellites and artificial intelligence creates more accurate wildfire maps for the public and firefighters.

The image illustrates the level of precision possible with a new fire imaging tool called RADRFIRE created by PNNL. (PNNL image)

Could RADRFIRE help with fire prevention? One area of ​​research is the use of LIDAR images, which can capture information about the nature and volume of vegetation in an area. The data could highlight areas with dense and at risk trees and shrubs that could be well suited for specific controlled burns or thinning to reduce available wildfire fuel.

This is modeling, AI, very mathematical stuff. But what about the human and environmental impacts of this work? “We are all absolutely driven by this. It is not unusual during these events where we all work 17, 18 hours a day. Nobody complains because we are all driven to try to do better, to do social good, “said Coleman. “For everyone on the team, this is personally important.”

Techno