Mark Williams and Bill Baker stand amid ponderosa pines in the mountains west of Fort Collins, Colo., holding a copy of a 19th century land survey. They're looking for a small pile of rocks with three notches on the east side, indicating that a General Land Office surveyor stopped here to describe the forest. Surveyors noted many things, says Baker, a professor at the University of Wyoming; they even discovered a gallows where two men had been hung. But generally the reports focus on forest type and structure -- less dramatic, perhaps, but more useful for modern researchers reconstructing historic Western forests and their fire patterns.

Over the past five years, Williams and Baker compiled thousands of hand-written descriptions and combined them with tree-ring data from the lines the surveyors walked. What they found surprised them. In each of their study areas -- mixed conifer and ponderosa forests in northern Arizona, Colorado's Front Range and eastern Oregon -- dense thickets of spindly trees and severe crown fires were common even before European settlement. In fact, the two scientists argue that the severity of many recent megafires, like Arizona's 2002 Rodeo-Chediski fire, which burned 190,000 acres, is actually pretty normal.

If that sounds counterintuitive, it is. Conventional wildfire wisdom is generally the opposite. Many scientists say that dry Western forests were once open and park-like, with large, widely spaced trees and little undergrowth.  Now, however, due to fire suppression and logging practices, they've become overgrown with small trees and shrubs. The result is that frequent low-severity fires have been replaced by a new era of megafires that are hotter and more severe than ever before.

That's true in some parts of the West, say Baker and Williams, a recent Ph.D. student, but not everywhere; many dry forests throughout the region historically were more dense and prone to severe fires. They also disagree with the idea that thinning and prescribed burns can prevent such fires.  That kind of treatment, applied in the wrong places, is not only misguided, they say, but could do more harm than good.

Naturally, those strong statements have met equally strong criticism from many pre-eminent fire ecologists. But Baker and Williams are not the first researchers to complicate fire ecology in the West. Unfortunately, the nuances in their and other scientists' perspectives are often oversimplified in the media and in policy-making, with damaging results. "A set of laws, policies and initiatives that aim to uniformly reduce fuels and fire severity is likely to (have) adverse effects on biological diversity," wrote Baker and Williams in  their recent paper in Global Ecology and Biogeography.

"It's very important that we take a more regional geographic approach and not apply what we know from one system to another," says Rosemary Sherriff, an associate professor of geography at California's Humboldt State University, whose work corroborates some of the Wyoming researchers' findings. But "it's hard to get that across, because the idea of park-like ponderosa pine is widespread."

One of the main researchers behind the traditional view is ecologist Wally Covington. When he began working at Northern Arizona University in 1975, many of the landscapes he encountered were choked with trees that had sprung up during decades of fire suppression. Covington wondered how the forests looked before widespread settlement.  His study results strengthened earlier findings that, in the past, Southwest ponderosa pine forests were kept open by frequent surface fires.

Subsequent recommendations to remove small trees and reintroduce surface fires became the basis for policies like the 2002 Healthy Forests Initiative, attracting the support of both lawmakers and environmentalists, since they would heal damaged ecosystems while reducing the risk of catastrophic fire.

The "Southwest model" gained support from other fire ecologists and began to be used as an explanation for forest problems elsewhere in the West -- even areas with vastly different historical fire regimes. Foresters in places like eastern Oregon, Montana and California's Sierra Nevada began thinning to make forests more park-like and reduce the risk of severe fires -- in the process jeopardizing wildlife such as black-backed woodpeckers and Kirtland's warblers, whose habitat is found in burn patches. Further research would reveal the situation's complexity; in reality, both surface and severe crown fires played an important role in many landscapes.

For example, the Sierra Nevada's ponderosa pine and mixed conifer forests are open and park-like on dry, south-facing slopes, with densely packed trees in creek beds and on northern aspects. Despite this variety, Malcolm North, a scientist with the U.S. Forest Service's Pacific Southwest Research Station, found in 2008 that fuel management officers were applying the Southwest model to an extreme, thinning the landscape into an "asbestos forest," rows of evenly spaced trees with no understory.