The character of change: New pathways in global change science
With a front-page headline on June 12, 2005, USA TODAY proclaimed, “The debate is over: Global warming is happening.” It was one sign of a profound shift in American attitudes over the last several years. Instead of arguing over whether the climate is actually warming or whether humans have anything to do with it, the nation’s political and economic leaders have begun to focus on the magnitude of the threat and how best to respond to it.
NCAR plays a unique and vital role in scientific and public discussions of climate change. With NSF support, the center has cultivated one of the world’s largest concentrations of researchers who specialize in climate analysis. These scientists and their university collaborators have produced hundreds of landmark studies published in peer-reviewed journals. Experts from NCAR—backed up by the center’s commitment to free and open dialogue—appear regularly in national and world media, where they join colleagues elsewhere in translating research of unusual complexity and importance into terms accessible to the public at large.
Close to 20 of the center’s physical and social scientists served as authors and coordinators on the long-awaited Fourth Assessment Report from the Intergovernmental Panel on Climate Change. The 2007 IPCC report—the world’s most definitive survey of climate change impacts, projections, and potential solutions—gave new visibility to NCAR’s work. The assessment itself drew heavily on the sophisticated skills of the NCAR-based Community Climate System Model. Supported by NSF and the U.S. Department of Energy, the CCSM has been developed in partnership with universities and federal laboratories and is freely available to all interested users. All told, the CCSM generated some 100 trillion bytes of data for the IPCC, the largest single contribution from any model.
What about water?
Despite the common shorthand of “global warming,” many researchers point to how climate change will affect the world’s water resources. “Global average temperature is an important abstraction,” says hydrologist and NCAR deputy director Larry Winter, “but water is what affects people the most. What comes out of your tap—that’s the real thing.”
Until recent years, global models disagreed on whether some regions would turn wetter or drier. The improved and enlarged team of models used in the 2007 IPCC report shows more consensus. In general, precipitation should increase north of latitude 40° north, while subtropical deserts push their way poleward.
One area of particular concern is the U.S. Southwest, where epic multidecadal drought tore at the fabric of indigenous life centuries ago. The last few years have seen catastrophic bouts of drought and wildfire across the region: in Los Angeles, 2006–07 was the driest winter ever recorded. A 2007 study led by Richard Seager (Columbia University) and coauthored by Jian Lu depicts an increasing risk of multiyear drought in the Southwest. Out of 19 models considered, all but one showed drying across the region, with an average regional decline in runoff of around 15% by 2021–2040 compared to 1950–2000. According to Seager, “The American West could be in for a future in which the climate is more arid than at any time since the advent of European settlement.”
Poring over 130 years of global data, NCAR’s Aiguo Dai and colleagues have found ample evidence of drying landscapes. Based on soil-moisture indices, they have shown that the percentage of Earth’s land area experiencing drought jumped from a range of 10–15% in the early 1970s to about 30% by 2002. “Global climate models predict increased drying over most land areas during their warm season. Our analyses suggest that this drying may have already begun,” says Dai.
Thunder, lightning, and modeling
For some parts of the world, the prognosis for moisture remains fuzzy. Most global models can’t yet track individual showers and thunderstorms (convection) because of the high computing cost for the dense three-dimensional framework needed to pinpoint and follow each storm cell. Instead, the models approximate the behavior of convection within larger grid boxes, typically about 100 by 100 kilometers (60 x 60 miles) horizontally. For modelers, it’s a painful compromise.
“We know that the global-scale models don’t handle convection very well,” says Roy Rasmussen, leader of the NCAR program on climate and the water cycle. In work that fed into the 2007 IPCC report, Rasmussen and colleagues found that the regions where models disagree most strongly on
how precipitation will change tend to be the same areas where seasonal convection rolls through on a near-daily basis. These populous regions include the central and eastern United States, much of India, the Amazon delta in Brazil, and Africa’s Sahel.
A multidisciplinary group at NCAR is working on ways to better approximate convection in global models. One common weakness is the lack of a simulated warm-air layer; in the real world, such a layer often suppresses thunderstorms till late in the day. Global climate models, in contrast, tend to produce large areas of gentler drizzle earlier in the day. “As soon as you get any instability in a climate model, it tends to start raining everywhere,” says Rasmussen.
In 2007 NCAR’s Mitchell Moncrieff and colleagues introduced an improved way to approximate, or parameterize, convection in models used for global weather prediction and climate research. Now being tested, the new strategy approximates the larger, longer-lived, and more-organized storm clusters known as mesoscale convective systems, as well as depicting the more common types of showers and thunderstorms traditionally represented in such models.
There’s also excitement about the NCAR-based Nested Regional Climate Model. The NRCM takes a standard global model and implants a finer-scale mesh of 36 km (22 mi) across the tropics and down to 4 km (2.5 mi) in and near Indonesia, where the effects of intense convection can reverberate far and wide. These computational telescopes show promise in replicating the subtleties of convective systems that regularly march across the tropics (see graphic). Once testing is complete, the NRCM will become an NSF-supported community resource.
A pole transformed: The Arctic’s accelerating melt
Like a slow, frigid heartbeat, ice forms across the Arctic Ocean each winter and recedes each summer. In terms of area covered, more than half of the ice each March is gone by September. Even then, the Arctic retains enough multiyear ice—averaging about 3 meters (10 feet) deep—to support polar bears and other ice-dependent life in much of the region.
However, the Arctic’s vast freeze-thaw cycle is shrinking in scope. Continued warming over the next few decades could bring an end to the ocean’s summer ice pack by late in the 21st century, according to the latest IPCC analysis and the 2005 Arctic Climate Impact Assessment. An ice-free Arctic in summer could not only imperil some native creatures but also bring down the curtain on many indigenous ways of life.
This bleak future may arrive even sooner than we thought, according to NCAR’s Marika Holland. By the 2040s, she calculates, late summers may find the Arctic virtually free of ice. “Our research suggests that the decrease over the next few decades could be far more dramatic than anything that has happened so far,” Holland says.
This finding emerged from a major improvement to the CCSM. Until recently, global climate models were hobbled by crude representations of the ebb and flow of sea ice. But over several years, a group led by Cecilia Bitz (University of Washington) and Elizabeth Hunke and William Lipscomb (Los Alamos National Laboratory) created a sea-ice component for the CCSM that stacks up well against its peers. “It’s the best sea-ice representation in a global climate model,” according to Bruno Tremblay (McGill University).
With the new sea-ice model in place, Holland and colleagues took the exhaustive CCSM simulations of 21st-century climate carried out for the IPCC (see above) and examined how Arctic ice might change. At first, the model shows a continuation of the gradual decline now under way. The pace picks up around the 2020s, when a rapid infusion of warm water from the North Atlantic melts large quantities of Arctic ice from below. This kicks a natural process into overdrive. As the ice retreats, sunlight that once was reflected to space enters the newly exposed ocean, which is much darker and more absorbent than ice. In turn, the warmed water melts even more ice. Within a decade, the CCSM shows September sea ice plummeting from about 80% of modern-day coverage to about 20%.
As ominous as this picture looks, recent events support it. A 2007 study led by Julienne Stroeve (National Snow and Ice Data Center) found that, from 1953 to 2006, the Arctic’s summer ice declined about 50% more quickly than shown by computer models as they attempt to recreate the period. “This suggests that current model projections may in fact provide a conservative estimate of future Arctic change,” says Stroeve.
The character of change | Top of the world | Shaping the atmosphere | Models and molecules |