Firefighters at the Ranch Fire in California
Firefighters stand silhouetted by the flames of the Ranch Fire, part of the Mendocino Complex fires in California. Credit: AP Photo/Noah Berger

Across the western United States, more than 100 fires are burning through tinder-dry forests. In California alone, one of the Mendocino Complex fires, the Ranch Fire, is the state’s largest on record, and the Carr Fire may go down as the state’s sixth most destructive fire by number of structures burned.

The model is a weather and smoke forecasting tool that helps communities downwind cope with soot-filled skies.

Even in places far from the flames, residents can feel the sting of wildfire smoke: In Utah, air quality is reported to be at “unhealthy” levels in some areas due to smoke, disrupting tourism and athletic events and leading to increases in hospital visits.

As the wildfire season continues to blaze, scientists at the National Weather Service (NWS) have been turning to satellites and numerical models to aid firefighters on the ground. One of their latest tools was developed by NWS’s parent body, the National Oceanic and Atmospheric Administration (NOAA). The model, from scientists in the Office of Oceanic and Atmospheric Research’s Global Systems Division, is a weather and smoke forecasting tool that helps communities downwind cope with soot-filled skies.

Eos spoke with one the people behind these tools: Andy Edman, chief of the Science and Technology Infusion Division of NWS’s western region. The interview below has been condensed and lightly edited for flow and grammar.

Eos: What is your role at the National Weather Service?

“There has been a strong message from both fire managers and community managers, such as mayors, that we need better forecasts.”

Andy Edman: I work in the western region headquarters here in Salt Lake City, Utah. Our group is responsible for the western third of the U.S. As researchers and academics come up with new ideas, my role is to take a hard look at them, and if it’s appropriate, to infuse the new research into the services that we provide.

Eos: How do you help communities fight fires?

Edman: Over the last several years, there has been a strong message from both fire managers and community managers, such as mayors, that we need better forecasts. If we wait till the fire is happening, it’s too late to react. A city needs time to get ready.

A view of California wildfire smoke seen from the Space Station
On 3 August 2018 aboard the International Space Station, European astronaut Alexander Gerst captured this image of smoke billowing from the Carr and Ferguson wildfires in California. Credit: NASA

Eos: And that’s where the NWS comes in?

Edman: NWS is responsible for providing watches, warnings, and forecasts of the weather and hydrological events. Wildfires are particularly driven by the weather, so we provide a number of different services. We provide forecasts to both state and federal land agencies, saying, for example, “Here’s what the weather is,” or pointing out if conditions are ripe for a fire. The agencies can adjust their tactics accordingly.

One of the recent trends in the firefighting community is that they’re getting much better about moving resources in anticipation of fire potential. If we’ve been having monsoon rains in Arizona or Utah, the agencies watching wildfire risk will start moving fire crews and repositioning them to, say, northern California, based on the forecast. About a week out, NWS will point to a region that’s going to be hot, dry, and windy, and fire managers will begin to plan and pre-position resources into the general area before a fire even gets started. It takes time to move crews and equipment.

Eos: The Carr Fire in California is now 71% contained, but it left in its wake a thousand buildings destroyed and resulted in at least eight deaths. How did the NWS assist the agencies fighting the Carr Fire?

Edman: A week or two before the Carr Fire started, the NWS started highlighting to state and federal agencies in northern California that the atmosphere was warming up and that it was going to get spectacularly warm and dry there.

Once the fire got started, we sent briefings whenever the fire crews needed them. Knowing something about the wind direction and speed, and how stable the atmosphere is, makes a big difference. If the air is unstable, it can help support an updraft on the flames and you’ll see more dangerous fire activity. We’ve also had an incident meteorologist, called an IMET, on the Carr Fire since almost day one.

Fire growth map of the California Carr Fire
A National Weather Service map of fire growth for the Carr Fire on 8 August 2018. Incident meteorologists give fire managers these maps several times a day to position firefighting crews on the ground. The area shaded yellow is the burned region, and the colored circles show active fires burning in the past 48 hours. The data come from NASA’s Terra satellite. Credit: NWS

Eos: Has firefighting changed over the years, given that NWS now provides current and expected ground conditions and wind speeds?

Edman: What’s changed in the last 30 years is that the fire managers can be much smarter about where they put the crews. When I started at NWS, we had some incidents when we had fires burn over the fire crews, and we lost people. That doesn’t happen very often these days. We’re better at anticipating where fire activity is likely to increase and positioning the fire crews accordingly. The fire tactics are smarter.

Eos: Where do you get your data to advise fire crews? From planes? Satellites?

Edman: It’s a combination of observations, such as satellites, radar, and surface observations, as well as numerical modeling.

Eos: You’ve been experimenting with a new model to predict where wildfire smoke will go. Even though it is still in its testing stages, the forecasting model, called the High-Resolution Rapid Refresh–Smoke (HRRR-Smoke), has been providing fire managers and communities with valuable insights. What does it do?

Edman: The model shows where the high levels of smoke concentrations are, and where they’re headed toward. We start with an initial state of the atmosphere, and from there the model makes a forecast for the next 36 hours.

“In the communities close to the fires, smoke is a real health hazard.”

Eos: Other than making the sky hazy, what are the risks from wildfire smoke?

Edman: It impacts where helicopters and airplanes can fly, because they have certain restrictions about when they can fly when the visibility gets too low, so they can’t do target drops or move fire crews around.

In the communities close to the fires, smoke is a real health hazard. For example, Yosemite National Park was closed for almost 3 weeks. The fire was physically outside of the park, but the smoke was blowing into the park. And the emissions were so high that park officials said that it was unhealthy for people to be there.

Eos: How does HRRR-Smoke work? Where does HRRR-Smoke get the latest data about active wildfires?

Edman: The HRRR model receives information about fire location and intensity from several polar orbiter satellites, the Joint Polar Satellite System (JPSS) and Moderate Resolution Imaging Spectro-radiometer (MODIS). The satellites sense the heat signature of the fires using infrared light, and from that we get the fire location and intensity.

Once we get that information, the next step is to translate those data into an estimate of how much smoke is coming off the fires. That information is used by the high resolution HRRR meteorological model to forecast where the smoke will go.

Eos: Who has access to the HRRR-Smoke model output? Can we see it, or is it just for the fire managers?

Edman: We give it to the fire crews, and our western offices have been including it in our routine forecasts. Community leaders are paying attention to it. For example, there’s a big bike race in Utah that’s sort of the local version of the Tour de France. The organizers are worried about the smoke, so the NWS has been giving the forecasts to the organizers.

Eos: You’ve said that HRRR-Smoke is still R&D. What could it do better?

Edman: It’s very much a work in progress. Every part of it needs to get better. We still struggle to model small fires and brand-new fires, because it takes time for the satellites to pick those up.

Eos: What do we gain from this new technology?

“We can show you a pretty picture of a hurricane, but the first thing you’re going to ask me is if it’s going to hit your house.”

Edman: This is a great example of where we’re going with science. It’s no longer good enough to just observe something. For example, we can show you a pretty picture of a hurricane, but the first thing you’re going to ask me is if it’s going to hit your house.

Eos: Does the Science and Technology Infusion Division have anything else new in the pipeline?

Edman: The NOAA GOES-17 satellite was launched about 5 months ago. Once that’s operational, we want to explore what we could do with those images, too. What we have currently is GOES-15, but it is the older generation. It’s getting to be close to 9 years old. Satellites are a little like cars; every generation gets better and better. GOES-17 can take imagery updates as quickly as every minute, whereas the other satellite can do it only every 5 minutes.

—Jenessa Duncombe (@JenessaDuncombe), News Writing and Production Intern


Duncombe, J. (2018), How forecasting models are changing the way we fight fires, Eos, 99, Published on 17 August 2018.

Text © 2018. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.