NOAA’s new AI wildfire detection system

A new artificial intelligence program developed by the National Oceanic and Atmospheric Administration (NOAA) aims to identify wildfires as small as an acre by analysing images from weather satellites orbiting approximately 22,000 miles above Earth.

As reported by Phys.org, the AI, called the Next Generation Fire System, processes vast amounts of satellite data, detecting heat from fires smaller than a football field and flagging potential new fires for human verification.

This system, tested in Boulder, promises to reduce the time between fire detection and response, a critical factor in fire containment.

Mike Pavolonis, NOAA Satellites’ Wildland Fire Program manager, explained the advantage: “While humans are great at detecting a new fire from satellite images, they can’t process the data as quickly and easily as the AI program.”

The system underwent its first significant test run with fire weather forecasters, state wildfire managers, and researchers using simulated fire scenarios.

Real-time application and success of the Next Generation Fire System

The Next Generation Fire System has been online in an experimental form for about a year, but agencies have now started using it in real time.

The system provides GPS locations of suspected fires, which are more accurate than traditional methods of locating smoke sources.

It also offers information about fuels in the area and potential weather conditions that could affect fire behaviour.

Zach Tolby, manager of NOAA’s Fire Weather Testbed, highlighted the tool’s importance for remote areas: “In populated areas, 911 gets called pretty quickly when there is new smoke. But that’s not the case in more rural areas.”

The program’s advanced detection capabilities are crucial as wildfires in the West have become larger and more devastating due to drought, climate change, and decades of fire suppression.

Broader AI applications and funding in wildfire management

NOAA’s artificial intelligence initiative is one of several projects leveraging AI to detect and combat wildfires.

In Colorado, multiple governments and companies, including Xcel Energy Colorado and the Aspen Fire Department, use AI-driven systems like those provided by Pano AI to monitor for smoke.

State lawmakers in Colorado attempted to allocate $2 million for AI camera systems in 2023, but the bill did not pass.

However, the federal government has invested heavily in new technologies for wildfire management, with the Fire Weather Testbed and the Next Generation Fire System funded through a $5 billion allocation in the Bipartisan Infrastructure Law.

NOAA Administrator Rick Spinrad commented on the impact of rising temperatures and frequent droughts: “Rising temperatures, declining snowpack and frequent droughts are all leading to a dramatic surge in wildfire frequency and severity across the United States.”

He added that the investments aim to develop tools to improve forecast accuracy and assist local officials in decision-making.

Continued advancements in wildfire detection technology

NOAA has long used weather satellites for wildfire detection, but recent technological advancements have significantly enhanced their utility.

The organisation is also working on an AI program to detect and predict lightning strikes, which can potentially start wildfires based on weather conditions and other factors.

Pavolonis expressed optimism about the progress made: “The rate at which progress has been made on this has been pretty exciting.”