Phil Rasch is a thoughtful guy with unruly hair and a quick smile who's at home in the clouds -- pondering the chemistry behind cloud formation and translating it into atmospheric models.

As the first chief climate scientist for Pacific Northwest National Labs, he's also one of the key climate specialists the world is relying on as it decides whether to combat climate change.

That puts him and his research findings at the center of a world drama as the United Nations Climate Change Conference in Copenhagen, Denmark, proceeds.

Rasch, 57, is convinced we need to reduce greenhouse gas emissions to keep global temperatures in a safe range.

In 2007, he signed the Bali Climate Declaration with 200 other climate scientists, urging world leaders to act quickly to keep a lid on carbon emissions. He's tinkered with the idea of reducing warming by producing sulfate particles in the sky.

But the work of climate scientists is under fire as never before.

Polls indicate fewer Americans in the past year believe the evidence behind global warming. Among Republicans and independents, the decline has been particularly steep.

And more than a thousand e-mails hacked from the Climatic Research Unit of England's University of East Anglia have exposed scientists to allegations of gaming results to favor a global warming hypothesis. The center's director, Phil Jones, stepped down last week in the wake of the "Climategate" scandal.

Rasch has escaped that storm. But he agrees the controversy comes at a bad time.

"I think the accusations will be resolved with time," he said. "But it won't happen by Copenhagen."

In the bigger picture, Rasch's work, focused on clouds and climate models, also targets two of the most hotly debated aspects of the global warming issue.

Clouds, it turns out, could increase or decrease as the planet warms, and their presence could boost global warming or reduce it.

It depends where they lie in the atmosphere. It depends on the size of their water droplets and ice particles. It depends on how they interact with airborne particles, or "aerosols," that could rise or fall in the future, from sulfate spewed by coal plants to the minerals that drift when farmers plow their land.

Climate models, based on thousands of calculations cranked through far-flung supercomputers, are behind U.N. projections of a 4- to 10-degree Fahrenheit increase in the globe's average surface temperature by 2100 if we continue to rely heavily on fossil fuels.

The 2007 Bali declaration from climate scientists urged keeping warming below 3.6 degrees Fahrenheit. Above that level, Rasch says, we could hit "tipping points" that accelerate warming, including sea ice disappearing and frozen Arctic methane being released as the ground that holds it warms.

Increases in carbon emissions are "like being on a teeter- totter," Rasch said. "Putting a small weight on one side or the other can tip the balance. That's the thing to worry about."

Rasch grew up mostly in Seattle, getting sold on deep science as a University of Washington undergraduate, when he helped study ice movements in the Arctic by floating around on an ice sheet for three months.

As a postdoctoral student in the early 1980s, he studied atmospheric models, discovering that the simplifying assumptions made in the models were swamping the real-world physics behind clouds.

Rasch worked for 28 years at the National Center for Atmospheric Research in Boulder, Colo. The models he helped develop have been used in assessments by the U.N.'s Intergovernmental Panel on Climate Change -- the target of global warming critics worldwide.

From his sunny office at the Department of Energy's National Laboratory complex in Richland, he's now trying to improve the models with results from the lab's cloud and airborne particulate research.

That research includes monitoring clouds in airplanes and mountaintops and creating simulated clouds in earthbound laboratories. It also includes helping to adjust the about 1 million lines of computer code, calculations and assumptions behind the Community Atmospheric Model that he and about 20 other scientists have built over two decades.

The latest Intergovernmental Panel on Climate Change report, in 2007, tapped 23 climate models to reach its findings. It concluded that warming of the climate is "unequivocal," based on temperature records, rising sea levels and "widespread" melting of snow and ice.

Among its other "robust" findings: Manmade greenhouse gas emissions rose by 70 percent between 1970 and 2004. And continued emissions at or above current rates will cause further warming and changes to the 21st century climate that would "very likely" be larger than those observed in the 20th century.

Those potential changes include rising sea levels, more extreme weather and reduced snowpack and water supplies in some regions.

Rasch stresses that the IPCC work includes extensive critiques, with every criticism tracked and addressed.

"This is a far more stringent process than any other scientific endeavor," he said. "The more you look at it, the more you see that it's a trustable process."

The models do a good job of gauging basic temperature patterns over ocean and land -- a crucial variable for global warming, Rasch says.

They're less solid when it comes to other changes, including the influence of clouds. The 2007 IPCC report pegged cloud feedbacks on the climate system and the impacts of sulfur and other airborne "aerosols" among "key uncertainties."

"You can state with absolute certainty if nothing else changed and you increase carbon dioxide the atmosphere will warm," Rasch said. "The issue is really that other things do change as well, and you have to factor them in."

Researchers test the models against historical climate data to see if they spit out credible results, and each of Rasch's papers ends with critiques of remaining weaknesses in the model. That's fodder for critics. It's also one of the basic tenets of science to identify holes in a hypothesis, Rasch says.

The models aren't flawless, he says, but they've improved dramatically.

"We know our (climate) models are imperfect," Rasch said. "It's possible we're all wrong. But the models are all pointing in one direction. The prudent thing to do is to assume that they're right but continue to test them."

Another challenge: The complexity of climate scientists' work makes gauging the truth awfully tough for lay people.

Piling an economic crisis on top of that head-scratching complexity hasn't helped when it comes to making global warming a priority.

Neither has the tougher questioning of the general climate scientists' consensus, from a smaller cohort of skeptical scientists and from pundits who allege that climate scientists exaggerate threats to rake in research money.

"I don't know how to deal with it, to tell you the truth," Rasch said.

"You can always say that the people you've deputized to take care of a problem are all in league with each other to get more research dollars," Rasch said. "To me, they're just doing their jobs."