Tree rings reveal increased fire risk for southwestern US

Historical record points to climate patterns that could prime the region for an intense fire season.
Hawk flying over burnt landscape in New Mexico

New Mexico's Las Conchas wildfire in 2011 burned so hot in places that scientists don't know whether the forests there will recover.Credit: Eddie Moore/Albuquerque Journal/ZUMA Wire

Ellis Margolis’s desk is covered in slices of wood. They’re scattered around his computer, piled in cardboard boxes and stacked on shelves. The tree rings and burn scars imprinted in the wood help the ecologist and other scientists predict the potential for future fires in the southwestern United States by understanding the blazes of the past.

This year, as Margolis and his colleagues at the US Geological Survey (USGS) in Santa Fe, New Mexico, go into the field to expand their dataset, the work feels like a race against time. Snowpack is dangerously low in many parts of Arizona and New Mexico, setting the stage for exceptionally dry conditions that fuel these conflagrations. The biggest wildfires in New Mexico's history have erupted in just the last decade, torching hundreds of thousands of hectares across the state, including in the mountains Margolis studies. “The lack of snow is scary,” he says. “We’re set up for a big fire year if things don't change dramatically.”

Storms that could deliver game-changing amounts of precipitation are unlikely, according to a 6 February federal water-supply forecast. And preliminary results from the network of sampling sites across two of New Mexico’s mountain ranges suggest that future blazes — even as soon as this summer — could be bigger than recent megafires.

Emerging patterns

Since the late 1970s, researchers have used tree rings and fire scars to reconstruct the fire history of an area and to understand how climate drives conflagrations. Tree growth rings vary in width with annual precipitation, providing a record of past climates. And when blazes burn a tree without killing it, they leave scars that can be dated along with the rings.

Initially, researchers simply tried to understand how frequently fires had burned individual stands of trees, says Tom Swetnam, a fire ecologist at the University of Arizona in Tucson who is now based in New Mexico. Land-management agencies such as the US National Park Service and the US Forest Service had followed strict fire-suppression policies for decades, but they were beginning to recognize fire’s ecological benefits. The agencies wanted to know how fire had behaved historically, so they could use it as a tool to promote forest health.

As Swetnam and others built fire chronologies, patterns emerged. Forests in Arizona, New Mexico, west Texas and northern Mexico all tended to burn in the same years, for instance. And Swetnam eventually linked active fire years to ocean-circulation cycles in the Pacific Ocean that dried out the Southwest1.

Multiple studies on mid-elevation forests of ponderosa pine (Pinus ponderosa) in Arizona and New Mexico found2 that before widespread fire suppression, low-intensity fires had burned through these iconic forests roughly every decade. Fire helped to maintain the open structure of ponderosa stands, and without it the forests grew thick with trees. This made them more vulnerable to big, hot fires.

One such blaze ignited in New Mexico’s Jemez Mountains in June 2011, when a tree fell on a power line. Dubbed the Las Conchas fire, it torched about 63,000 hectares, making it the largest conflagration in New Mexico’s recorded history at the time. It burned incredibly hot, and killed so many trees in certain areas that scientists are unsure if the forest will ever grow back.

In some ways, the Las Conchas fire was an outlier. Its severe temperatures were almost certainly unusual, says Craig Allen, a USGS fire ecologist based in the Jemez Mountains, especially in areas where ponderosa pines were incinerated.

Yet, in other ways, fires on the scale of Las Conchas may be consistent with historical norms. For instance, Margolis says, researchers don’t have a good handle on whether the size of the blaze was truly exceptional.

A matter of time

This is one of the questions that Margolis has pursued by systematically sampling trees across the Jemez Mountains, and in expanding a similar network in the Sangre de Cristo Mountains outside Santa Fe. “It’s so basic,” he says, “but it takes a lot of data to get there.” Although his analysis of the Jemez data isn’t complete, Margolis sees strong evidence that fires as big as Las Conchas, or even twice its size, have occurred for centuries. The implication: “We should be more freaked out that the fires can get even bigger,” he says.

This year, it’s the potential for fire in the Sangre de Cristo Mountains that really worries Margolis. Many of the area’s forests haven’t burned in more than 100 years, an unnaturally long dormant period, according to fire histories he’s reconstructed. That means the mountains are stocked with fuel.

Widespread fires of the past burned on the heels of extremely dry winters that followed two to three wet winters. That’s exactly the pattern New Mexico is currently experiencing, because 2016 and 2017 were relatively wet.

“If we had a Las Conchas fire outside of Santa Fe, it would be devastating,” Margolis says. The fire itself could threaten life and property. The loss of vegetation would leave the area vulnerable to post-fire flooding that could wipe out roads and clog vital water infrastructure with debris. “We’re sitting on this powder keg,” he says.

Nature 554, 283-284 (2018)

doi: 10.1038/d41586-018-01686-y
Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up


  1. 1.

    Swetnam, T. W. & Betancourt, J. L. Science 249, 1017–1020 (1990).

  2. 2.

    Margolis, E. Q., Huffman, D. W. & Iñiguez, J. M. Southwestern Mixed-Conifer Forests: Evaluating Reference Conditions to Guide Ecological Restoration Treatments. Ecological Restoration Institute Working Paper No. 28 (2013).

Download references