After crunching ice-age climate numbers, Oregon researchers and colleagues from Harvard, Princeton, Cornell and Barcelona came up with two encouraging conclusions about future global warming:
The planet appears less sensitive to carbon dioxide changes than expected, their study says, so extreme temperature increases in the near future appear highly unlikely.
And future warming may also be less than predicted by the
in 2007, particularly at the upper end of the "likely" range.
"I think we should be worried, but we should not be desperate," said Andreas Schmittner, an
researcher and lead author of the study, published online today by the journal
. "It's not already too late to do something. We still have time to figure out a solution."
It may seem strange that the frosty climate 21,000 years ago, the rough timing of the last glacial maximum, could inform today's contentious global warming debate.
But the distant past has some key advantages over the present: The climate "signal" was unmistakable -- it was an ice age after all.
And the climate over the huge time frames involved was relatively stable, unlike the short period when humans have tracked variables such as temperature, precipitation and atmospheric dust.
The study, funded by the
paleoclimate program, drew on the known extent of ice sheets in the past and levels of carbon dioxide, methane and nitrous oxide taken from air trapped in Antarctic ice cores.
Researchers also mapped ice-age temperatures based on pollen levels on land and concentrations of temperature-sensitive microorganisms in the ocean.
Schmittner and colleagues then ran a climate computer model at different "climate sensitivities" -- the climate's reaction to greenhouse gas levels -- to see which sensitivities best pinpointed actual ice-age temperatures.
Their conclusion: The climate appears less sensitive to greenhouse gases than prior estimates. Based on the computer runs, doubling carbon concentrations would likely increase the world's average temperature from 3.1 to 4.7 degrees Fahrenheit over preindustrial levels, the study predicts.
That's lower than the IPCC's likely temperature range for a doubling of carbon dioxide: 3.6 to 8.1 degrees. And it's a far cry from increases of up to 18 degrees held out as low probability possibilities.
The study also concludes that increases greater than 11 degrees from a doubling of C02 "should be assigned near-zero probability." Computer runs using such severe climate sensitivity modeled the globe as entirely covered in ice during the ice age, the study said. Schmittner said the actual number was closer to 10 percent.
The IPCC anticipates doubled C02 levels by 2100 under its most eco-friendly vision of the future global economy and by 2050 in its most fossil-fuel dependent scenario.
In a "perspective" article, also published online today by Science, two researchers from the
not involved with the study praised it for tapping more ice age data than previous research, but pointed to significant caveats.
Schmittner's study used only one climate model, they noted. Other models do a better job of pinning down atmospheric variables that affect climate, such as cloud formation.
Significant holes also remain in some of the ice age data. And using secondhand climate information -- such as pollen levels -- isn't as certain as direct observations in modern times. Given those uncertainties, they said, "a firm upper boundary (of potential temperature change) is still elusive."
Schmittner agreed that future studies should use more models to better pin down the results -- the study used the one available to the researchers. But he said he's confident the study's "much lower" upper limits on temperature increases are robust and consistent with previous ice-age climate reconstructions.
Global warming critics have already cited the study's publicly circulated abstract to criticize the IPCC and predictions of harm from global warming. Schmittner said he's not in that camp.
Even the lower temperature increases predicted in the study could do serious damage, he said. The yardstick for the IPCC and the study -- global average surface temperature -- also masks the potential for larger increases in some regions.
And the study indicated that relatively modest swings in ocean temperatures can cause dramatic climate changes, Schmittner said, including ice formation and land temperatures.
"I don't think (the study) rules out melting of ice sheets or permafrost," he said. "The cautionary tale here is that very small changes in the ocean can correspond to very large changes over land."
Eight researchers co-authored the study, "Climate Sensitivity Estimated from Temperature Reconstructions of the Last Glacial Maximum." Oregon scientists involved include Schmittner, Peter Clark and Alan Mix of Oregon State and Patrick Bartlein of the
.
-- Scott Learn