The Backbone of California’s ALERTWildfire Camera Network Is UC San Diego

An interdisciplinary team of faculty and students is combating the increasing threat of wildfires.

Image courtesy of Scripps Institute of Oceanography.

ALERTWildfire camera network. (Image courtesy of Scripps Institute of Oceanography.)

Wildfires are a growing problem in California. Over the last 10 years, the state has seen record-breaking fires, and there is no sign that the situation will improve. This has made fire management a priority for California and much of the West Coast.

The ALERTWildfire camera network is a new tool for wildfire management. The system was codeveloped by the University of California, San Diego (UC San Diego) and built in partnership with the University of Nevada, Reno (UN Reno) and the University of Oregon. It is based on the highly successful ALERT Tahoe system, which was deployed in 2010.

ALERT Tahoe was a pilot project in the Tahoe Lake region created by the Nevada Seismological Laboratory (NSL) at UN Reno. That network was originally designed to detect earthquakes. However, its application to forest fires soon became apparent.

The Fire Guards from Meadow Vista, Calif. Image courtesy of the Nevada Seismological Laboratory.

The Fire Guards from Meadow Vista, Calif. (Image courtesy of the Nevada Seismological Laboratory.)

A team of students from Meadow Vista, Calif., who called themselves the Fire Guards, suggested that a network of wirelessly connected cameras in the forest could monitor forest fires. The team envisioned this network of cameras being connected to social media to better engage the public. Their insight won them the Innovate Award in 2009 at the Children’s Climate Action in Copenhagen.

The NSL embraced the Fire Guard’s idea. A network of wireless near-infrared HD cameras was installed throughout the Tahoe Lake region revolutionizing fire protection in the area.

ALERTWildfire

Tahoe ALERT spread out of the Lake Tahoe region into northern Nevada. The success of the program from 2014 to 2016 caught the eye of the Bureau of Land Management across the West Coast. Several electricity companies were also interested in the new fire management technology. Around this time, UC San Diego and the University of Oregon joined the project and ALERTWildfire was born.

The camera network quickly demonstrated its value when combating wildfires. Neal Driscoll, the director of the ALERTWildfire program at UC San Diego, explained in a news release, “These cameras save critical time by allowing rapid confirmation of 911 calls and accurate location of new fires using the ALERTWildfire web-based interface, time that would otherwise be spent sending engines to mountaintops or launching aircraft to confirm fire ignition and location.”

Wildfires start small, so the quicker firefighters can get to a new fire, the more manageable the situation will be. The camera network is a game changer, allowing firefighters to take an offensive position instead of a defensive one.

Tony Mecham, the CAL FIRE San Diego Country Chief, said in a UC San Diego news release, “The cameras have given us real-time situational awareness and have allowed us to make informed decisions much quicker than we used to. It used to take 20 to 30 minutes for our fire ground commanders to get to fires and make decisions, and now with the cameras we are reacting within seconds of the first report. That extra time is significant when it comes to moving resources or needing to start evacuations. It’s making a difference. I can’t even put into words how important those first few minutes are.”

Each camera can see for 60 miles on a clear day and, thanks to the near-infrared night vision, 120 miles on a clear night. They monitor the area 24 hours a day and complete a 360-degree sweep of their location about every two minutes. The cameras can also be controlled remotely, allowing them to zoom, tilt and pan for better viewing.

But there is more to this system behind the scenes. UN Reno and UC San Diego’s Jacobs School of Engineering manage the cyberinfrastructure of the cameras. This is no small task—one year of camera data exceeds one million gigabytes of data. The amount of raw data is only expected to grow as the network of cameras expands. This requires next-level data processing and wildfire visualization techniques.

The tools that the school is building to handle all this data will be open source. This will allow the software to be utilized by other disciplines, extending the utility of the system. Ilkay Altintas, an investigator on the project, hopes that the software will benefit the larger educational community by making “it possible for students from the high school to graduate level to participate in uploading their own data logging, data processing or data-driven alerts.”

Still from an ALERTWildfire camera courtesy of the Nevada Seismological Laboratory.

Still from an ALERTWildfire camera. (Image courtesy of the Nevada Seismological Laboratory.)

WIFIRE Integration with ALERTWildfire

UC San Diego is focusing on utilizing real-time knowledge about wildfires. Toward that goal, the university has used funding from a National Science Foundation grant to create the WIFIRE Lab project. WIFIRE will integrate ALERTWildfire real-time data with predictive fire behavior modeling.

WIFIRE uses an interdisciplinary approach that brings together researchers from the San Diego Supercomputer Center, the Qualcomm Institute, the Scripps Institution of Oceanography, and the Mechanical and Aerospace Engineering department at Jacobs School of Engineering.

Currently, the project wants to extend ALERTWildfire with a fleet of drones. This will require overcoming many engineering and computing challenges. Jacobs School of Engineering is designing the drones. The fleet will provide not only imaging, but also real-time sensory data such as temperature, humidity, wind, and air quality information. All this data will provide early warning data points for predictive models.

Nikolay Atanasov, an electrical and computer engineering professor at UC San Diego, is working with the departments of Engineering and Computer Science to develop algorithms that will better coordinate the drones. The students in Atanasov’s lab must tackle issues such as autonomous navigation, cooperative mapping, collision avoidance, and autonomous recharging of the drones.

Even without drones, WIFIRE is revolutionizing fire management. Now, when a bush fire is reported, WIFIRE retrieves data from weather stations about the current conditions and associates that data with topographical information about vegetation and other fuel sources in the area to predict the fire spread. All of this happens within minutes.

Within 30 minutes, firefighters are on scene with water-dropping helicopters. As the planes fly over the fire, the information they collect about the fire is used to confirm and correct the model created by WIFIRE. The maps plotting the predicted fire path, which are updated every couple of hours, are used to plan evacuations.

Downwind Effects of Wildfires

The research does not stop at fire management. Rosana Aguilera, a postdoctoral researcher at the Scripps Institution of Oceanography, is researching air quality. This has created a joint venture between Scripps and the School of Medicine. The project includes Alexander Gershunov, a climate scientist, and Tarik Benmarhnia, an epidemiologist jointly appointed to Scripps and the School of Medicine.

Their research is exploring how fine particulate matter (PM2.5) created by wildfire smoke affects the public’s health. It is well known that the fine particulates in smoke can exacerbate a range of health problems. The research will analyze the impact this has on different regions.

Aguilera explained in a news release, “Most studies looking at health effects of wildfires offer global estimates for a wildfire event and/or region.… We are looking at both the temporal and spatial distribution of hospital admissions within Southern California.”

ALERTWildfire Is Making a Difference

Fire management has come a long way. In 2017, there were not many cameras installed throughout the West Coast. Since that time, the fires in the region have only intensified. The network of cameras has now grown to 610 cameras and is playing a critical role in fire management.

The cameras give firefighters a new tool that is “worth its weight in gold,” according to Ben Nicholls, California Department of Forestry and Fire Protection chief in Sonoma County. It allows firefighters to respond to wildfires quicker than ever before. For example, the California Department of Forestry and Fire Protection was able to respond to the Kincade Fire 11 minutes after it ignited.

Real-time data has been instrumental in saving lives. This is the motivation behind the creation of ALERTWildfire. Driscoll, its director, explained, “We’re saving lives. That’s the No. 1 thing.… The situational awareness allows first responders to scale resources and make important decisions in a timely fashion.”