When Fire Returns to Forests That Forgot How to Burn

THE ANCIENT FORESTS of the Pacific Northwest have a problem: they’re too old in all the wrong places. Across 24 million acres of federal land spanning western Oregon, Washington and California, the towering stands that once thrived with regular low-severity fires now face a paradox: the very forests that historically shrugged off flames are today’s kindling.

A new analysis reveals the geography of this vulnerability in stark detail. Three-quarters of mature and old-growth forests now at highest risk of catastrophic wildfire sit in areas where, historically, fire passed through frequently but gently. These weren’t forests that burned rarely and intensely – quite the opposite. They were shaped by flames that arrived every 5 to 50 years, clearing underbrush whilst barely singeing the big trees.

Fire exclusion changed all that. When Indigenous peoples were forcibly removed from their lands in the 1850s, millennia of careful fire stewardship ended abruptly. Then came 1910 and the Great Fire – a conflagration that destroyed several towns and scorched an area roughly the size of Connecticut across Idaho, Montana, Washington and British Columbia. Federal policy responded by putting out every wildfire possible. Over the next century, the forests transformed.

Bruno Aparicio, a postdoctoral researcher at Oregon State University who led the new study, has been mapping exactly what’s at stake. Working with satellite data, fire behavior simulations and historical records, his team identified which forests are most exposed – and where topography might offer natural refuges. The findings paint a picture of landscapes holding their breath.

Consider the Shasta-Trinity National Forest in California’s Klamath Mountains, which faces the highest exposure among old-growth forests – roughly 800 hectares burning at high intensity each year, representing 1% of its ancient stands annually. The Okanogan-Wenatchee in Washington follows close behind. For mature forests, the pattern intensifies: Okanogan-Wenatchee sees about 2,500 hectares exposed yearly.

The Klamath Mountains emerge as a particular flashpoint, harbouring 37% of the region’s mature and old-growth forests whilst experiencing the highest annual burn area at severe intensity. It’s a concentration of both ecological treasure and combustible risk.

What makes a forest vulnerable isn’t just age or density, though. Forest type matters enormously. White fir and grand fir forests – which represent what ecologists call “potential natural vegetation” in areas that historically supported fire-resistant species like ponderosa pine – now face the greatest total exposure. These are landscapes where fire exclusion allowed shade-tolerant, fire-intolerant species to fill in beneath the canopy. The resulting dense, continuous fuels create conditions ripe for crown fires that race through treetops.

Yet there’s nuance in the numbers. Subalpine fir forests, whilst relatively rare, show the highest percentage of their total area exposed annually – over 1% each year. These high-elevation forests near the timberline have always experienced infrequent but severe fires. Climate change is now pushing fire deeper into these zones.

“Now, as wildfire activity intensifies under climate change, understanding where and why mature and old-growth forests are most vulnerable is essential,” Aparicio says. His analysis provides something managers have long needed: a spatial framework showing where intervention might help and where nature’s own defences might suffice.

Those natural defences – what researchers call fire refugia – represent pockets of landscape less prone to crown fire. Think of them as the slow lanes of wildfire, places where terrain, vegetation and proximity to water create conditions that resist stand-replacing burns. The concept has spawned an entire field of refugia science.

The study examined two types of refugia: topo-climatic (based on landforms and weather patterns) and holistic (incorporating fuels and vegetation as well). When combined, these refugia could potentially reduce severe fire risk by up to 21% across the study area. But the effect varies wildly by location. The Siuslaw National Forest, with 85% of its old-growth in fire refugia zones, enjoys substantial natural protection. The Fremont-Winema, by contrast, has just 4% of its ancient forests in refugial areas.

Meg Krawchuk, who supervised the research at Oregon State, emphasizes the carbon dimension. Since 2000, federal lands in the region have lost a net 2.6 million acres of mature forest and 700,000 acres of old-growth – first to logging, now increasingly to wildfire. These forests store vast amounts of carbon, and when they burn hot, that carbon heads skyward.

The team estimates about 300,000 tons of carbon will be emitted annually from stand-replacing fires in these old forests. More than half comes from just two forests: Okanogan-Wenatchee and Shasta-Trinity. The Klamath Mountains account for 54% of expected emissions regionwide.

“Protecting mature and old-growth forests isn’t just about preserving the past — it’s a key strategy for climate mitigation, ecosystem resilience and long-term forest stewardship,” Krawchuk says. Her team’s analysis reveals that fire refugia could cut those carbon emissions by up to 21% – significant, but hardly a complete solution.

The research underscores an uncomfortable truth about fire exclusion’s legacy. Frequent low-severity fire wasn’t a threat to these forests; it was their maintenance regime. Ponderosa pines, Douglas-firs and other fire-adapted species evolved thick bark and self-pruning branches specifically to survive regular burns. The fires killed competing seedlings, cleared fuel, and maintained the open, park-like structure that made subsequent fires mild.

A century without fire allowed shade-tolerant species – grand fir, white fir – to establish dense thickets beneath the old pines. Fuel accumulated on forest floors. Stand structure shifted from scattered large trees to continuous canopies. When fire returned under today’s hotter, drier climate, it found conditions primed for catastrophe.

The irony is sharp: forests that historically burned frequently now face the highest risk precisely because they haven’t burned. Meanwhile, forests in wetter zones that always experienced infrequent, severe fires continue in that pattern. The Coast Range and North Cascades, characterized historically by mixed-severity regimes, show more alignment between past and present fire behavior.

Aparicio’s analysis offers managers a roadmap, highlighting 11 of 16 national forests where more than half the high-intensity burn area occurs in historically frequent-fire zones. These are the places where restoration – thinning, prescribed fire – could have the greatest effect. “Our work can help prioritize stewardship actions,” Krawchuk notes. The alternative is watching these forests convert to something else entirely.

The study appears in Natural Hazards, joining a growing body of work documenting how fire regimes have shifted across the western United States. The findings matter beyond the Pacific Northwest – they’re a preview of what happens when you suppress a natural process for a century, then climate change arrives to collect the bill.

Some managers are already acting, using prescribed fire to reduce fuel loads and restore forest structure. Others are identifying fire refugia as conservation priorities, recognizing these landscape pockets as critical for maintaining biodiversity through the coming decades of intensifying wildfire. The challenge is scale: 7 million acres of mature and old-growth forest across the region is a lot of ground to cover.

There’s also the question of what restoration means for truly ancient stands. You can thin a crowded forest back toward historical conditions, but can you safely light fires beneath 400-year-old trees that haven’t seen flames in a century? The risks of getting it wrong – killing the very giants you’re trying to protect – make managers cautious.

Still, the arithmetic is stark. At current rates of exposure, these forests face ongoing attrition. Some will survive in refugia. Some will burn and regenerate. Others will transition to different ecosystems entirely, potentially locking in a new fire regime for centuries to come.

The researchers emphasize their modeling captures broad patterns but can’t predict which specific stands will burn in any given year. Fire behavior models, whilst sophisticated, remain probabilistic tools. Fire refugia mapping likewise represents likelihood, not certainty. Nature retains her capacity for surprise.

Yet patterns matter when you’re managing millions of acres. Knowing the Klamath Mountains face disproportionate risk, that subalpine forests show increasing vulnerability, that fire refugia are unevenly distributed – this knowledge shapes strategy even when individual fire seasons remain unpredictable.

“Overall, fire exclusion has left historically fire-resistant forests disproportionately exposed to severe fire now,” Krawchuk says. “Our analysis can underpin the stewardship of mature and old-growth forests to promote their resistance and resilience.” It’s a measured statement that acknowledges both the problem and the possibility of solutions.

The forests of the Pacific Northwest have outlasted Ice Ages, volcanic eruptions and centuries of logging. Whether they can survive our fire paradox – too much protection followed by too little – remains an open question. But at least now we know where to look for answers.

study link: https://www.nature.com/articles/s44304-025-00160-w