Studies that found evidence droughts are growing more prevalent may have overestimated the global drying trend, new research suggests.

In its 2007 report, the Intergovernmental Panel on Climate Change cautioned that "more intense and wider droughts have been observed over wider areas since the 1970s, particularly in the tropics and subtropics," and drew a link among changing drought trends, increasing temperatures and decreasing precipitation.

But a study published yesterday in Nature suggests the IPCC, and the research it drew on, relied on a flawed index that may have overestimated drought activity over the past 60 years.

The work examines the Palmer Drought Severity Index, a formula devised in the 1960s to allocate drought aid to farmers. It estimates drought by calculating the difference between precipitation and soil moisture. Because it requires only basic data -- observations of temperature, precipitation and latitude -- the Palmer index has become a popular tool for scientists, who have used it to extend records of drought into the late 19th century.

Justin Sheffield, a hydroclimatologist at Princeton University, said that simplicity has a downside.

The Palmer index doesn't account for some key influences on evaporation, including radiation at the land surface, humidity and wind speed. That didn't matter for a long time, Sheffield said, because the major factors that affect evaporation -- temperature, radiation, humidity and wind speed -- were changing in concert. And that allowed the Palmer index to produce realistic estimates of drought using only temperature data, assuming that as temperature rises, evaporation increases.

But over the last several decades, the relationship among the factors governing evaporation has begun to unravel.

Old index may be too simple

"We see wind speeds declining, and we see 'global dimming'" -- a drop in the amount of sunlight reaching the Earth's surface -- "which means radiation is going down, at least between the 1960s and 1990s," Sheffield said.

And that means the simplified math of the Palmer index is no longer accurate in many cases.

"What we have been doing is using a more physically realistic model of evaporation that includes changes in these other variables that are important," Sheffield said. "You get a different answer, and a more correct answer, because you are using better physics."

When Sheffield and his colleagues examined global drought severity from 1948 to 2008 using the more complex model, they found little change in global drought trends from 1948 to 2008.

"This study is a bit of a wake-up call that we need better observations, we need better data and we need better models to understand what's happening with drought and other features of the land hydrological cycle," Sheffield said. "It's where we live."

Other experts agreed. Piers Forster, a professor of physical climate change at the University of Leeds, said the research "shows that a much-used previous estimate based only on temperature is flawed."

"In reality, we don't have enough good-enough measurements to determine drought trends accurately," Forster said in a statement to the U.K. Science Media Centre.

Critic finds conclusions too sweeping

But even though the paper finds no global average increase in drought since 1950, scientists are more confident that droughts in Europe, Africa and China have become more severe since then, he said.

Sonia Seneviratne, a climate scientist at the Swiss Federal Institute of Technology in Zurich, said the new study suggests "high uncertainty" in drought trends globally over the past 60 years and "little evidence" of an increase in the area affected by drought.

Furthermore, its "findings imply that there is no necessary correlation between temperature changes and long-term drought variations, which should warn us against using any simplifications regarding their relationship," Seneviratne wrote in a commentary that was also published in Nature.

But Aiguo Dai, a professor at the State University of New York, Albany, said it is "premature to make such definite conclusions about recent drying trends."

In a recent study published in Geophysical Research Letters, Dai evaluated different formulas that attempt to account for evapotranspiration. He found that the Penman-Monteith equation, the formula favored by Sheffield, "only slightly reduces the drying trend" seen in the original Palmer index.

Dai also rejected the idea that Sheffield's conclusions, if true, would affect model projections of future drought using Palmer index values.