A 10-year timeout for global warming, study says

The German research effort is one of the first to attempt 10-year climate forecasts.

Global warming is taking a break that could last for another 10 years or so.

That's the latest word from a team of climate researchers in Germany. Global average temperatures should remain above normal, the team suggests. But additional warming – already on hold over the first seven years of this decade – is likely to remain that way for another decade. The reason? The team says it expects natural shifts in ocean circulation to affect temperatures in ways that temporarily out-wrestle the effects of rising greenhouse-gas emissions.

The forecast is "very bold," cautions Tom Delworth, a scientist at the National Oceanic and Atmospheric Administration's Geophysical Fluid Dynamics Laboratory at Princeton University. But, he adds, it represents the cutting edge of climate modeling. The German effort is one of the first widely published attempts to offer climate forecasts on time scales of a decade or so, rather than a century or more. The findings appear in Thursday's edition of Nature

Even without global warming, decades-long natural shifts in climate can have big social and economic effects. These changes are thought to drive highs and lows in the average intensity of a string of hurricane seasons or the recurrence of persistent periods of drought, for instance. That alone makes the effort to forecast them worthwhile, researchers say. But these shorter-term climate forecasts also can act as a more immediate reality check on century-scale climate projections, since the same computer models are being used for both tasks. And they have the potential to identify more clearly those cases where global warming is responsible for triggering a decade's climate patterns, rather than natural variability.

"These are nice first steps," Dr. Delworth says of the efforts so far.

The latest attempt comes from climate modelers at the Leibniz Institute for Marine Sciences in Keil, Germany, and the Max Planck Institute for Meteorology in Hamburg. But they used their global model in a slightly unorthodox way.

Climate models are bursting with equations that describe physical processes taking place in and among ice and snow, the ocean, and atmosphere. Typically, scientists plug in a small handful of outside "forcing" conditions – the amount of light the sun emits as it undergoes its own cycles, as well as levels of man-made and natural greenhouse gases and tiny particles called aerosols. Then they let the model run. In the process, the model recreates natural variations. But until now, these variations have often been considered "noise" that interferes with teasing out global warming's long-term trajectory.

But the German team was looking to more finely model that noise to forecast climate change on much shorter time scales. To do that, they borrowed a page from weather forecasters, who plug in measurements from weather balloons, satellites, and other sensor platforms as a starting point for their computer calculations.

For decadal-scale climate forecasts, however, the atmosphere changes far too rapidly to be of much use, some researchers say. Instead, the German team wanted to use slower, large-scale ocean circulation patterns and their interaction with the atmosphere. But such ocean measurements are few and far between. So the German team used ocean-surface temperature patterns as stand-ins. They used these temperatures over a wide swath of the planet as the kick-off point for their model, in addition to the external forcings they typically use. Then they let the model run with no more interventions.

The team found that the models' ability to predict changes improved significantly with the ocean-temperature measurements added. The result: Surface temperatures over much of Europe and North America are expected to cool slightly over the next decade. Meanwhile, ocean temperatures in the tropical Pacific will remain largely unchanged. On balance, that yields a period of relatively stable global average temperatures.

Last year, a group at Britain's Hadley Centre for Climate Prediction and Research issued their own initial decadal forecast. The group used both measured ocean and atmospheric conditions to kick off their model, in addition to what they termed "plausible changes" in future natural and man-made greenhouse gases, aerosols, and in sunlight. Their outlook: Natural variations will offset global warming to some extent, although the team expected climate to continue warming. At least half the years after 2009 are projected to exceed the warmest years currently on record.

Both groups acknowledge that a lot of work lies ahead before such models are ready for prime time. For one thing, researchers would like to be able to use ocean temperature and other measurements that reach as deep as 5,000 meters, Delworth says. Depending on how additional research plays out, the new approach could find its way into the next set of reports from the Intergovernmental Panel on Climate Change, due out in 2013.

For people interested in applying the results to practical planning problems, it's still too early to tell how useful these forecasts might be, notes Jonathan Overpeck, an atmospheric scientist and director of the Institute for the Study of Planet Earth at the University of Arizona in Tucson.

In the US Southwest, for instance, knowing that warming might level off over the next decade "would be pretty useful for people" if the forecasts were skillful enough. "But we really don't know how skillful these will prove to be. Smart stakeholders – particularly in the West, where high temperatures are causing problems with managing forests, with managing water resources, and plenty of other things – are going to hedge their bets."

Even so, these studies carry a couple of take-home points, he continues. First: Don't be surprised at cool periods interspersed among the warming; natural variability will still occur, even though superimposed on a broader warming trend. "The anthropogenic signal is clear," he says. "But there's also natural variability around the signal." And second: Even if the forecast for relatively stable temperatures holds up, "the warming could come back with a vengeance with a subsequent warming part" of the see-saw pattern, he says.