IoT Sensors Called to the Wildfire Front Line

The Department of Homeland Security explores ground-based sensors near wildfire spark points to speed alerts to first responders and state and local emergency management agencies.

The damages from fire seasons out west keep piling up as the dangerous fires spark earlier in the typical fire season, burn bigger, and continue longer. Unfortunately, the 2020 season is far from over. Since the beginning of the year, there have been more than 8,300 wildfires in California that have burned more than 4 million acres in the state and caused 31 fatalities, as of early October.

Experts predict that California, Oregon, Washington, and other western states will continue to face these battles each year. Knowing that fires have—and will be—a constant in these areas, organizations are tapping technologies like IoT to help lessen the destruction. The data from IoT sensors has the potential to signal early alerts and help create new predictive models that can save property, homes, and lives.

Homeland Security Enlists IoT for Wildfire Testbed

One of the interesting efforts underway is the Wildfire Sensor Technology program supported by the Department of Homeland Security (DHS) Science and Technology Directorate and the Smart Cities Internet of Things Innovation Lab (SCITI Lab). The two organizations have joined forces to accelerate the development of commercially viable solutions for state and local agencies. To develop the best possible technology-driven solutions for the job at hand, DPS and SCITI are applying design-thinking concepts and asking the groups that will use these solutions to provide feedback.

The intention is to have ground-based wildfire sensors combined with an IoT platform and data analytics that will enhance preparedness. The sensors will track smoke, humidity, air quality and more, as close as possible to the spark point, providing early warning, and hopefully, faster and safer responses.

The team has chosen four sensor technology partners and each is developing a prototype based on requirements the DHS created for combating wildfires that impact urban areas.

The four multi-modal sensor providers include:

Automated Real-time Alerts

The goals for the sensor companies are to focus on real-time and continuous identification of heat sources, smoke, particulates, organic compounds, gases, and chemicals to detect ignition location, track fire perimeter, track fire characteristics, and allow for geographically targeted notifications and warnings. The sensor data will be augmented with information from the National Oceanic and Atmospheric Administration and National Weather Service satellites and aerial surveillance systems.

In the first demonstration, the companies explored the use of visual detection using AI, particulate and smoke sensing, air quality and infrared sensors to inform real-time alerting to first responders and communities.

Based on the feedback and assessments about the prototypes, FEMA and state and local fire services requested additional predictive and forecasting tools to help them better understand what the fires will do next. AI, machine learning, and predictive analytics will be added into the mix to understand how the fire is behaving and where it may move. This blend of technologies will give agencies both detecting, forecasting, and tracking tools to contain the wildfires.

The end goal is to create a market-ready solution that is usable, affordable, and scalable for DHS, first responders, and other stakeholders. The expectation is that these solutions will join the National Weather Services and its GOES satellite system, thermal camera networks, and Doppler radar networks that are already working at full pace to better manage and detect wildfires.