OREGON STATE UNIVERSITY

Hot, wet future bodes ill for forest fire danger

09/03/2002

CORVALLIS, Ore. - Global climate changes will probably cause most of the American West to become significantly warmer and wetter in the coming century, experts say, producing an increase in vegetation growth that could lead to repeated, catastrophic fires greater than any in recent history.

Increases in precipitation, higher temperatures and rising carbon dioxide levels in the atmosphere will combine to spur vegetation growth and add even further to the existing fuel loads caused by decades of fire suppression, say scientists at Oregon State University and the U.S. Forest Service.

Unless there are significant changes in land management practices, the amount of trees, shrubs and other vegetation could increase beyond current levels. Droughts or heat waves could then lead to levels of wildfire larger than most observed since European settlement.

These findings are based on a growing body of published research and increasingly improved computer models that predict both future climate scenarios and their impacts on forests and other terrestrial ecosystems, said Ronald Neilson, a courtesy professor at OSU and bioclimatologist with the U.S. Forest Service. However, these models do not consider land use changes that could mitigate or, in places, exacerbate direct and indirect impacts.

"The American West is going to get warmer and it's going to get wetter, generally leading to an expansion of forests, more vegetation overall and significantly more fire," Neilson said. "Our computer models combine what we now know about natural climate trends and variation, along with the climate changes caused by human impacts and industrialization, and they are becoming quite accurate."

Various "general circulation" models show global warming of 5-12 degrees Fahrenheit by about 75 years from now, and two of the more commonly cited models show precipitation increases in the U.S. of about 20-30 percent.

In those conditions, one computer model shows the amount of live vegetation doubling, or in some cases even tripling in large portions of California and the American Southwest by late in this century. Significant amounts of increased vegetation, often 10-50 percent, may occur in parts of Oregon and Idaho, with lesser amounts of vegetation growth and even some decreases in Washington state.

In the same time frame, the amounts of "biomass" consumed by fire will double or triple in large parts of Nevada and Southern California, and increase 50-100 percent in most of Oregon east of the Cascade Range.

"On the drier forests of the western interior, adding more precipitation to the system is just like pouring on gasoline," Neilson said. "You increase the fuel load, inevitably have dry years with hot summers, and you get catastrophic fires."

The future of fire in the western portions of Oregon and Washington is more variable and less clear, Neilson said.

"Increasing precipitation in what are already mostly wet forests may actually decrease fire risks, and that's what the models indicate for some areas," Neilson said. "But it's also clear that these climate trends will increase the amount of tree growth and understory vegetation to even higher levels than we now have, so one possibility is much less frequent, but highly destructive fires during very dry or hot periods."

And such dry or hot periods may come with increasing frequency, the climate models suggest.

There are several forces at work that scientists are just now coming to understand and be able to piece together in their computer simulations. One is the variation brought on by the "El Nino" and "La Nina" events, which are anomalies in currents of the Pacific Ocean that can have large, but short-term effects on weather around the world.

Then, there is the Pacific Decadal Oscillation, a 20-25 year cycle of climatic forces. And finally, there are the long-term changes being wrought by global warming or the "greenhouse effect."
"These short, medium and long-term climatic changes all interact with each other, sometimes offsetting impacts on temperature and precipitation, and sometimes compounding them," Neilson said. "It's complex, but we're rapidly improving our ability to forecast how they will combine to affect near-term climate, on a level of decades, years or even months in the future.

"But one thing that's already clear," he added, "is the cycle we're in right now could lead to some extremes of temperature and precipitation."

However, a key variable that the models do not consider is changes in human land management practices that could serve to either address the problem or change the issues involved, said Dominique Bachelet, an associate professor in the OSU Department of Bioengineering.

"We can't predict what land use changes may be caused by humans, or what types of forest management activities such as thinning may take place that would decrease fire risk," Bachelet said. "With increased population growth it's even possible that some areas we predict may burn in the future could become agricultural fields or urban areas."

One of the existing problems is the movement of people into forested and dryland rural areas, complicating forest and fire management issues, Bachelet said. Changes in land use regulations to address this problem, more tolerance of natural fire in dry forests, and more aggressive management of forest fuel loads would all change the equation for future forest and rangeland fire risks, she said.

And where properly used and controlled, Bachelet said, fire should also be looked at as an ally - it can be a key player in reducing fuel loads and restoring forest health.

Meanwhile, researchers are continuing their work to fine-tune the computer models upon which many of these scenarios are based.

One interesting observation that has just come to light is a clear signal of recent "fire suppression" efforts and a suggestion that nature may ultimately have its way despite any human efforts.

Some models have correlated climate predictions with fire trends as far back as the late 1800s. They are remarkably accurate in "predicting" which years had the worst fire seasons. But in recent decades, the amount of acres that burned were often less than the model predicted should have burned - unbeknown to the computer, the natural fires were being extinguished by human fire suppression activities.

In just the last few years, however, the rate of increase in acres burned in the U.S. significantly exceeded the rate the computer predicted.

"An obvious thing these models may be telling us is that you can put out fires for a while and suppress them below the natural amount they would have burned," Neilson said. "But then the fuel loads build up so high that you have fires far more catastrophic than you would have had under natural conditions, and there's almost nothing you can do to put them out."

Ultimately, Neilson said, the computer models should incorporate human activities, such as past fire suppression, fuel loads, or changes in land management approaches.

"There no doubt will be variables we'll need to re-examine and changes that will need to be made as we improve this process of climate prediction and its ecological effects," Neilson said. "But from a land management perspective, one thing that seems pretty clear is we will have to manage for a changing ecosystem, not for the status quo.

"Barring significant changes, the future of the American West appears to be one that is hotter, wetter and includes a lot of fire."