2. Ultimate cooker

3. See rising sea level

4. Know for sure?

5. What to do?

On average, 832,000 acres burn annually in U.S. national forests, but 2002 is proving to be one of the hottest years ever. USDA.

Truly warming?
So it looks like warming, and smells like global warming caused by human activities. Is it actually global warming?

Sadly, things get murky at this point, and climatologists cannot say for sure whether global warming is taking place. Climatology is a physical science, like chemistry and physics, and physical scientists like to test hypotheses by performing enough experiments to get a valid result.

In climatology, such a hypothesis might say, "Given the current rate of increase in greenhouse gases, what global average temperature can we expect in 50 years?"

An even better understanding might come from a controlled experiment: "Let's hold other greenhouse gases constant but double carbon dioxide, to see how carbon dioxide influences warming by itself."

These are great experiments, but you simply can't do them as you would do, say, an experiment in particle physics.

If you think of climate predictions as hypotheses, the only way to test them is to watch climate behave through the years.

"If you want to wait 100 years, I can tell you for sure" whether global warming is a fact, says Jonathan Foley, an environmental scientist at University of Wisconsin-Madison. "But by the time we have perfect evidence, global warming will be so severe that it will be impossible to undo."

Twice as nice with dice!
One way to understand the human effect on climate, says global warming expert Stephen Schneider, is to compare it to loaded dice. Although you never can predict the next roll, loading the dice skews the odds in favor of a certain number.

And loading the climate dice is exactly what we are doing as we pump billions of tons of greenhouse gases into the atmosphere each year, Schneider says.

Because it's impossible to do the experiments, it's impossible to know if the torrid pace of warming in recent years really signifies rapid warming. "Twenty years is not a long time," says Schneider, "and we don't have a pair of climate dice to roll 1,000 times to figure out if this is an accident."

A comparison of annual, average thermometer readings (red line) to four computer models of climate (black lines). The two lines are similar until 1960 or so. "Anomalies" are differences from baseline, 1880 to 1920. Charts from IPCC Climate Change 2001, Technical Summary, Working Group I, 2001. Source: p. 58.

Schneider, however, is convinced that human actions are the best explanation for the current heat-up. "I'd argue that global warming at the surface is a virtual fact. At a very high probability, way above 90 percent."

Still, he's willing to admit that he could be wrong. "Most people in climate science would argue that we are loading the dice, but to know for sure, you would have to run the experiment."

Can't compute
Climatologists have tried to sidestep the "no-experiment" limitation by running climate models on big computers. While the models predict much of the warming that's occurred in the atmosphere, they suffer from the garbage-in, garbage-out problem.

Bad data will produce errors (and don't forget that acquiring data is a big problem in climate studies).

And no matter how much models are refined, it's not possible, even in theory, to include every relevant factor. Sure, aerosols were factored in after Mt. Pinatubo, but how, for example, would we get data on these critical scientific, political and economic questions:

Will warming trigger a quick change in ocean circulation, drastically altering regional climate? If so, when?

Will warming trigger a geologic change releasing enormous amounts of the potent greenhouse gas methane? When?

What will greenhouse gas emissions be in a century?

If rising sea levels push millions of people from their homes, where will they go, and how will that affect future warming trends?

Must compute
At some point, Foley argues, the discussion must move from the scientific realm, with its excruciating standards of proof, to the political realm, where "Regular folks who need a decision now can look at what's a reasonable degree of evidence."

Once you add in the human effects on global warming, predictions match measurements. The drop in 1992 was caused by reflective particles from Mount Pinatubo. IPCC, p. 58.

While climatologists argue the fine points, he asks, "Where is there a reasonable doubt about global warming? Every piece of evidence says the world is getting warmer and CO2 is rising. Every law of physics says that increasing CO2 will trap heat. We don't have 100 percent statistical proof, but we've got a smoking gun, a motive and a body. What else does a regular person need to make a decision?"

In law and medicine, decisions are routinely made without perfect evidence, Foley continues. "Scientists have the luxury of living in an ivory tower, of not making decisions. We have to stay honest to the science, but we also need to recognize that regular people need to move forward without perfect data."

So what should we do now?

There are 1 2 3 4 5 pages in this feature.
Bibliography | Credits | Feedback | Search

©2002, University of Wisconsin, Board of Regents.