In summer 2003, more than 52,000 Europeans died from heat-related ills, 30,000 in France alone, during an unrelenting heat wave that featured temperatures 6.5 degrees Fahrenheit (3.6 degrees Celsius) higher than normal. Crops also suffered, with corn production down by 30 percent and wheat by 21 percent, among other foodstuffs. And a similar hot spell in Ukraine in 1972 led to a wheat shortage that prompted that staple's prices to more than triple by 1974. But even without record-breaking heat, recent years have seen food riots from Bangladesh to Haiti as world agriculture was pushed to the breaking point by a combination of greater demand for food, biofuels and poor weather.

Such disruptions in the world's food supply may become even more the norm by the end of this century, according to a new analysis published today in Science. Climate modeler David Battisti of the University of Washington in Seattle and food security expert Rosamond Naylor of Stanford University used the results of 23 climate models to determine that there is a more than 90 percent chance—in other words, it is very likely—that the lowest growing season temperatures in the tropics and subtropics by the end of the century will be higher than the highest temperatures at present.

That area includes the southern U.S., Central America, southern Europe, central Asia, northern Australia and all of Africa, according to Battisti. "Although it had not been calculated before," he says, "it was not surprising to find that for most of the tropics and subtropics, the future summer temperatures would be out of bounds compared to what we have ever experienced."

Planning meetings for the Global Seed Vault in Norway spawned the idea of looking at average summer temperatures, which climate models can project relatively reliably and which have a large impact on crop yields—between 2.5 and 16 percent less wheat, corn, soy or other crops are produced for every 1.8–degree F (1–degree C) rise. "The impacts we will see on yield, combined with a growing population that depends greatly on agriculture for food and income, will demand a profound level of adaptation, which might include moving hundreds of millions of people," Battisti says.

According to the projections, the temperate zones, like most of the continental U.S., will also be affected. "By the end of the century, however, the seasonal growing temperature is likely to exceed the hottest season on record in temperate countries (equivalent to what France experienced in 2003), and the future for agriculture in these regions will become equally daunting," the researchers wrote.

To date, concerns about climate change's impact on agriculture have focused on drought—another likely outcome of warming world. As a result, plant scientists have researched ways to develop drought-resistant strains of various crops, such as a variety of corn that agriculture giant Monsanto Company and chemical company BASF have submitted to the U.S. Food and Drug Administration for approval.

Funding for such agricultural research in general has been dwindling in recent years, according to the Washington, D.C.–based International Food Policy Research Institute.

"Current levels of agricultural research are not adequate to help developing countries to adapt to climate change," says IFPRI water resources expert Claudia Ringler. But "the dire picture in the paper assumes no adaptation. Adaptation will certainly happen. Farmers in Russia will change the time of wheat planting [for example], and may switch to a different crop if the hot summer of today becomes the norm of the future."

Battisti and Naylor, however, assumed greenhouse gas emissions lower than the present output and the fact that more carbon dioxide (CO2), the most common greenhouse gas, will boost plant growth may not help. The forecasted CO2 boost—as much as 10 percent—in crop growth will be more than offset by the 20 to 40 percent drop due to higher temperatures alone—and will be further exacerbated by any drying, Battisti warns.

That means the future of agriculture as the climate changes could be even worse than this prediction—and that's before taking into account other factors such as the effect of pests.

"We want to look at the impact of climate change on the distribution of pests and pathogens that affect crops," Battisti says, "starting with maize in Africa." With three billion people living in the affected regions, at least a billion of whom are already malnourished, figuring out how to adapt agriculture to global warming couldn't be more urgent.