The Dust Bowl — the seven-year drought that devastated large swathes of US prairie land in the 1930s — was the worst prolonged environmental disaster recorded for the country. But a study of the American Southwest's more distant climatic past reveals that the catastrophic drought was a mere dry spell compared to the 'mega-droughts' that were recurring long before humans began to settle the continent.

The findings, reported in a paper in Nature¹ this week, add to concerns that the already arid region might face quasi-permanent drought conditions as climate continues to warm.

The team, led by Peter Fawcett, a climate scientist at the University of New Mexico in Albuquerque, reconstructed the region's climate history using geochemical indicators from an 82-metre-long lake sediment core from the Valles Caldera in northern New Mexico. Analysis of climate and vegetation proxies, such as pollen and carbon-isotope ratios, suggests that the Southwest experienced abrupt and surprisingly pronounced climate shifts during warm periods of the Pleistocene, including transitions to extended dry periods that lasted for hundreds or even thousands of years.

Reliving the past

If today's climate repeated past patterns, the southwestern United States might move into a wetter and cooler phase. Such a transition happened at one point during the so-called Marine Isotope Stage (MIS) 11, an interglacial period around 400,000 years ago that shows some striking parallels with the Holocene, our current warm period. This seems to have roughly advanced to the point at which the climate in MIS 11 began to switch to a less arid one.

Earth's orbit and axal tilt during the unusually long MIS 11 stage was similar to orbital conditions during the Holocene, which scientists think will last longer than most Pleistocene warm periods.

But for all the similarities, the climate during MIS 11 was unperturbed by human activity. This time around, rising greenhouse-gas concentrations driven by human activity will very likely override any natural cooling trend. Scientists fear that the Southwestern climate may in fact switch to an extended dry mode such as the ones that occurred during particularly warm Pleistocene periods.

"We won't know for sure if it happens again until we get there," says Fawcett. "But we are certainly increasing the possibility of crossing a critical threshold to severe and lasting drought conditions."

Sudden shifts in carbon isotopes and lowered total organic carbon in the sediment record suggest that grasses and shrubs that depend mostly on summer rain died out during extended Pleistocene droughts. This is surprising, says Fawcett, because summer monsoon rainfall was thought to become more intense in a warmer climate. That summer rain was in fact strongly reduced, or had almost stopped, suggests that regional climate patterns must have shifted radically when Pleistocene temperatures crossed a threshold.

"The scary thing is that we seem to be very close to this point again," he says.

A dry future

The Southwest has experienced significant reductions in rainfall during the last decade, causing freshwater reservoirs and groundwater to fall to unusually low levels. Colorado River flows recorded at Lees Ferry, Arizona, from 2000 to 2009 are the lowest on record.

Climate models suggest that the region will in future become even drier as atmospheric circulation patterns change and subtropical dry zones expand towards the poles²

"The drying we expect for the twenty-first century is entirely the result of increased greenhouse forcing," says Richard Seager, a climate researcher at the Lamont-Doherty Earth Observatory in Palisades, New York. "Any natural variations in orbital forcing and incoming sunlight will hardly have a noticeable role in the near future."

A 10–15% reduction in rainfall is enough to cause severe drought in the region, he says. Meanwhile, debate continues among scientists whether a transition to quasi-permanent dry conditions is imminent or already underway, and to what extent global warming has increased the risk of drought.

"A signal of anthropogenic drying is emerging, but it is still small," says Seager. "I'd expect that by mid-century the human signal will exceed the amplitude of natural climate variability. Then we can safely say that the Southwest has entered a new climate stage."