The Synoptic View of California’s Camp Fire: A Scorching Reality of Today’s Fires

February 19, 2019

The Synoptic View of California’s Camp Fire: A Scorching Reality of Today’s Fires

February 19, 2019

As 2018 closed, Discover Magazine blogger Tom Yulsman posted to his ImaGeo blog about the most gripping image he’d seen during the year. The “particularly compelling” one he chose was a Landsat 8 image of the Camp Fire in northern California.

L8 image of Camp Fire — The specific Landsat image of the Camp Fire that Yulsman selected was created by independent blogger and geojournalist Pierre Markuse, using Landsat 8 data collected on November 8, 2018.

Inception

The Landsat 8 data behind this image, which made repeated appearances in the media, told a striking story of both the extent and the speed of the blaze.

Landsat 8 acquired the image on Thursday, November 8 at 10:45 a.m. local time, a mere four hours after a small fire off Camp Creek Road had started the conflagration. It would become California’s deadliest fire, killing 86 people and burning more than 14,500 homes and businesses in the foothill town of Paradise.

EO Landsat image of the Camp Fire — NASA Earth Observatory provided a heavily used version of the Landsat 8 image of the Camp Fire in Northern California. USGS, universities, and environmental companies all provided versions of this image as well. The open Landsat data archive allowed anyone who wanted the data to get it. Image credit: Joshua Stevens, NASA Earth Observatory/Landsat

The Landsat image revealed that half of Paradise was already burning that morning. The fire’s extent was visible through the smoke because of data the satellite collects from wavelengths beyond human range of sight (namely the shortwave infrared (SWIR) and Thermal Infrared (TIR) regions).

Landsat 8 images of Camp Fire — Smoke obscures the ground in the visible color Landsat 8 image on the left (created using the red, green, and blue bands—bands 4,3,2). The image on the right was created using one of Landsat 8’s thermal infrared bands, one of its shortwave infrared bands, and its blue band—bands 10,6,4). The shortwave and thermal infrared data show the active fire front and hotspots below the smoke cover. Image credit: Matt Montanaro and Aaron Gerace/Landsat.

The Camp Fire burned more structures than any before in California’s history (18,804 in total when including support structures like sheds and garages). Flames consumed nine out of every ten homes in Paradise, in what is now recognized as the deadliest fire in America since 1918.

As Yulsman wrote in his initial blog post two days after the fire began:

“In all the many years that I’ve covered wildfire, I don’t believe I’ve encountered anything like what we’ve seen with the Camp Fire blazing in California’s northern Sierra Nevada mountains. What really shocked me was the speed with which this cataclysmic inferno progressed.”

At UC Berkeley, geography professor Jeff Chambers used this Landsat 8 image to calculate the speed of the fire. Given the known start time of 6:30 a.m. and the collect time of the satellite, he calculated that the fire had been traveling 3 mph during those first four hours, stoked by dry easterly winds tearing across California’s Sierra Nevada mountains from the warm deserts of Nevada. Winds speeds that day were 30 mph with gusts up to 50 mph.

Chambers and other fire specialists also used this Landsat data to verify the “leapfrog” behavior of the fire. Embers blown by the dry winds started spot fires along the western front, jumping 2.5 miles ahead of the main fire front. The Landsat image captures the fire behavior behind this devastating disaster.

As Chambers told Berkeley News, “It’s astonishing. You can see the fire burning right through the city.”

In nearby Chico, California, Zeke Lunder, a pyrogeographer and founder of Deer Creek GIS, used the Landsat image together with topographic data and road overlay information to create a number of 3D renderings of the fire.

3D render of Camp Fire — Lunder used the Landsat 8 image of the campfire together with topographic and road network data to create this image. This image and others created with Landsat 8 were used in public safety meetings and were featured in the Wildfire Today publication. Image credit: Zeke Lunder, Deer Creek GIS/Landsat/Google Earth

Lunder and his team used the Landsat image to create fire perimeter and progression maps. He used the maps in public meetings to, as he explained, “help people understand how the fire spread in its initial stages.”
Deer Creek GIS often uses Landsat to do regular landscape-scale fire hazard assessments, but the timing of this Landsat 8 image and what it showed was astonishing to even Lunder:

“We rarely are able to find images such as this, which capture the peak burning period of a major fire we are working. The size of the spot fires, distance the spot fires were ahead of the main fire, the hot spots—each a burning home—the width of the flaming front, and the fire’s impingement on all of the major evacuation routes, all tell a crazy story, capturing a remarkably dramatic and traumatic point in time.”

The town of Paradise is nestled into the foothills of the Sierras, situated on a ridge between two canyons. It’s a gorgeous geography, but one with limited escape routes. The Landsat 8 image captured how horrifyingly fast the fire spread, and why there was little or no time for evacuations. The Guardian interviewed 50 Camp Fire survivors from Paradise and their story paints a vivid picture of just how terrifying this fire was to live through.

It took more than 2,000 fire fighting personnel and nearly three weeks to finally quell the flames.

vegetation cover - Camp Fire — Map showing the mix of vegetation types that were within the Camp Fire perimeter. Image credit: Allie Weill, using CALVEG and CAL FIRE data.

Fueling the Fire

In interview after interview in the days after the fire started, fire researchers commented that the dry conditions accompanying our changing climate have shifted the nature of fires, increasing their ferocity.

Sprawling urban fires that once regularly plagued civilization were thought to be a thing the past—the Camp Fire let us know they are back.

Why was the Camp Fire so fierce, so aggressive? The drought conditions and the strong winds were factors, as was the “fuel,” i.e. what was burning.

After some mischaracterization, the Camp Fire was clarified as a wildland fire—a fire burning a mix of grasses, shrubby vegetation, and forest. It was this patchwork of vegetation that fueled the Camp Fire—a mix of vegetation that, after years of drought conditions, had become a dry fuel load.

As fire ecologist and science writer Allie Weill wrote for KQED Science prior to the Camp Fire, “Fuel Matters” and firefighters must pay close attention to land cover because it effects how a fire behaves.

Weill created a number of land cover maps for her article showing the different types of vegetation, or fuel, that was found where some of California’s worst fires were located. To create the maps she used open CALVEG data along with fire perimeter information from CAL FIRE, the California fire protection entity.

The CALVEG dataset is today derived from a mix of Landsat data, aerial data and field information, but the data set has a long association with Landsat.

Its acronym stands for Classification and Assessment with Landsat of Visible Ecological Groupings. The U.S. Forest Service created the initial dataset using first generation Landsat data in the 1970s.

Prevention and Preparedness in the New Age of Mega-Fires

fighting the Camp Fire — Firefighters attempting to hold the fire line during the Camp Fire. Photo credit: InciWeb

With a repeat cycle of every sixteen days it is rare for Landsat to image a fire within hours of its onset, as it did for the Camp Fire. But the mission’s long archive and well-calibrated data set makes it essential for fire management in other ways.

In addition to informing land cover data sets such as CALVEG, Landsat is integral to a number of fire management programs including LANDFIRE, an interagency endeavor that provides vegetation fuel data for the entire U.S.; Monitoring Trends in Burn Severity, or MTBS, a program that uses Landsat to map burn severity and fire extent, watching for fire behavior trends, and the Burned Area Emergency Response program, which puts teams on the ground armed with Landsat-based burn severity maps immediately following fires to aid in post-fire remediation efforts.

NASA uses Landsat alongside its other space-based Earth-observing resources to provide data to fire managers before, during, and after a fire as part of its Disasters Program. Landsat also helps inform NASA’s RECOVER program that provides online mapping tools with spatially relevant data that fire managers need.

Meanwhile, the European Space Agency is test-piloting a Fire Detection Service that processes Landsat data in real time, scanning for any fire signatures and alerting responders when a fire is detected.

As the Camp Fire has forced us to recognize, the trend towards bigger and faster wildfires cannot be ignored, especially as cities grow towards wildland areas.