The people of East Africa once again face a devastating drought this year: Crops wither and fail from Kenya to Ethiopia, livestock drop dead and famine spreads. Although, historically, such droughts are not uncommon in this region, their frequency seems to have increased in recent years, raising prices for staple foods, such as maize.

This scenario may simply be a taste of a world undergoing climate change in the mid–21st century, according to a new report from the International Food Policy Research Institute (IFPRI), a Washington, D.C.–based organization seeking an end to hunger and poverty through appropriate local, national and international agricultural policies. By IFPRI's estimate, 25 million more children will be malnourished in 2050 due to the impact of climate change on global agriculture.

"Higher temperatures and changes in precipitation result in pressure on yields from important crops in much of the world," says IFPRI agricultural economist Gerald Nelson, an author of the report, "Climate Change, Agriculture, and Food Security: Impacts and Costs of Adaptation to 2050". "Biological impacts on crop yields work through the economic system resulting in reduced production, higher crop and meat prices, and a reduction in cereal consumption. This reduction means reduced calorie intake and increased childhood malnutrition."

Nelson and his colleagues, working with funding from the World Bank and Asian Development Bank, estimated global agricultural impacts by pairing IFPRI's own economic models for crop yields with climate models for precipitation and temperature from the U.S. National Center for Atmospheric Research and Australia's Commonwealth Scientific and Industrial Research Organization. Assuming a world that is slow to adapt to climate change and focused on regional self-reliance, the researchers found that children in the developing world—which are the countries expected to provide the bulk of population growth to nine billion or more by mid-century—will be hardest hit.

"It's not economic development that matters in this case, it's the location on the surface of the Earth," Nelson notes. Without better crop varieties or other agricultural technology improvements, irrigated wheat yields, for example, will fall at least 20 percent by 2050 as a result of global warming, and south Asia as well as parts of sub-Saharan Africa will face the worst effects.

Even without climate change, population pressure alone will cause a spike in food prices without intervention, according to IFPRI's economic model. For example, without climate change, wheat prices might rise from $113 per metric ton in 2000 to $158 per metric ton in 2050—an increase of 39 percent. Similarly, rice prices would soar by 62 percent, maize by 63 percent. But factoring in climate change will boost wheat prices by at least 170 percent and rice by a minimum of 113 percent; the cost of maize will be at least 148 percent higher than at the turn of the century by mid-century.

Nor will the developed world go unscathed. Research published in the Proceedings of the National Academies of Science in August noted that corn, soybean and cotton yields in the U.S. will drop precipitously because of additional days where the temperature is above 30 degrees Celsius.

Part of the problem is that the benefits of better plant growth, thanks to higher carbon dioxide concentrations (plants use CO2 for photosynthesis) are more than offset by the impact of higher temperatures and differing precipitation. "If you grow a plant in a bell jar in a lab and increase the CO2 inside, the plants will perform better. [But] will those results translate into farmer's fields? Evidence that we've been getting from farmer's fields suggests perhaps not," Nelson says. And that means fewer calories per person would be available in 2050.

To prevent this agricultural crisis, Nelson estimates, would require an investment of at least $7 billion per year in the most affected countries in Asia, Africa and Latin America for increased agricultural research into, for example, drought-resistant crop varieties. "Crop and livestock productivity–enhancing research, including biotechnology, will be essential to help overcome stresses due to climate change," the report's authors wrote.

These areas will also need expanded rural road and irrigation infrastructure as well as improvements to the efficiency of that irrigation.

Climate change's glacial meltwaters will not aid such irrigation projects. "The glaciers, particularly in the Himalayas, may disappear and cause some of the major rivers to become much more variable, which will have negative effects on yields in south Asia," Nelson notes. At the same time, traditional seed varieties and livestock breeds that might provide a genetic resource to adapt to climate change are being lost.

Crop diseases and insect pests will also thrive in a hotter or more humid climate, and the report does not take into account issues such as current agricultural lands swamped by rising sea levels. "These are conservative estimates," Nelson adds. "Some elements we left out could make those numbers even higher."

Even those areas that will benefit from a changed climate, such as a potential expansion in regional climates amenable to certain crops in Canada, for example, will not solve the problem. "The problem is you'd have to grow corn on some pretty rocky soils," Nelson explains. "It's not clear that you'd get more production even if climate favors them."

And expanding agriculture to feed more people may simply exacerbate climate change. Deforestation, largely driven by conversion to cropland, accounts for roughly 16 percent of global emissions of the carbon dioxide warming the atmosphere.

There is hope, of course. IFPRI's fellows in the Consultative Group on International Agricultural Research system are developing, for example, drought-tolerant or heat-resistant varieties of staple crops such as wheat and rice. And various efforts such as the Millennium Villages in East Africa may speed adaptation. Already, the Kenyan village of Sauri has boosted maize yields with the help of an influx of donor cash. And some Indian farmers in the state of Bihar have begun to plant hybrid rice strains because they are drought-tolerant and can be planted on lands that were previously difficult to successfully cultivate.

"Agriculture is the sector most likely to be affected by changes in climate of all sectors of society," Nelson adds. "Investment will not guarantee that all negative impacts can be overcome, but business as usual will guarantee disastrous consequences for the human race."