Seasonally adjusted
SINCE time immemorial, farmers have planted their crops according to the seasons. That is what my forefathers have been doing, says Mohammad Ilisasuddin in Shibganj, in northern Bangladesh, but now the weather does not seem right for what we have done traditionally. Seasonal planting is useless, agrees Florence Madamu, a smallholder in Bulirehe, in western Uganda. The sun is prolonged until the end of September and whenever it rains, it rains so heavily it destroys all our crops. Oxfam, a British charity, has compiled a litany of laments by poor farmers. John Magrath, a researcher, says they all say similar things: moderate, temperate seasons are shrinking...rainy seasons are shorter and more violent...making it more difficult to grow crops [and] difficult for them to know when best to plant. As the earth warms up, many have feared that farmers will pay a high price. But working out who will pay, how, and where is tricky. Higher temperatures might turn arid shrub lands into deserts while improving the growing season in colder steppes. Global warming could produce more evaporation from plants, and more rain, which would benefit some places, while hurting others. In theory extra carbon dioxide in the atmosphere should help plants grow faster, though whether this actually happens may also depend on the amount of nitrogen in the soil. In the most comprehensive effort* so far to think these questions through, the International Food Policy Research Institute, a think-tank in Washington, DC, has reached some sobering conclusions. In parts of the developing world some crop yields in 2050 could be only half of their 2000 levels. Irrigation may not help: climate change will hit irrigated systems harder than rain-fed ones. And the hope that gainers from climate change will outweigh losers looks vain: the damage from higher temperatures and erratic rainfall will be too big. In its forecast IFPRI started with the A2 scenario of the Intergovernmental Panel on Climate Change. This is the second-gloomiest of six IPCC scenarios (it assumes the world will be releasing roughly twice as much CO2 in 2050 as it does now) and says the oceans' surface temperature will rise by around 1.6C by 2050. However, this says nothing about the temperature and rainfall patterns that would result on farmland. To forecast those, IFPRI fed the IPCC assumptions into two climate-change models, one run by America's National Centre for Atmospheric Research (NCAR), the other by Australia's Commonwealth Scientific and Industrial Research Organisation (CSIRO). These gave different descriptions of the world in 2050. NCAR thinks the climate would be hotter and wetter, with rainfall about 10% heavier than now. The CSIRO forecasts that there would be 2% more rain. There were big regional disparities, too: CSIRO forecast the sharpest increases in temperature in southern Africa; NCAR sees Russia and Canada heating up more. To take account of the differences IFPRI fed both forecasts into its own computer, which describes how every agricultural region and, in some places, practically every farm, responds to changes in temperature and rainfall. The results varied less than the assumptions. In developing countries, IFPRI found, irrigated wheat in 2050 would yield 34% less than in 2000, using NCAR data; and 28% less going by CSIRO figures. For irrigated rice, the declines would be 19% and 14% (see chart). These falls are large but not unlikely: scientists in South Africa recently said the region could see a 50% fall in cereals productivity by 2080. Bad though they are, the average declines hide even more disturbing variations. Latin America comes out of the exercise relatively well: the yields of its main crops are expected to fall by only a few percent. China's farming may also be more resilient than it sometimes appears. But South Asia, the world's most heavily populated region, looks vulnerable: IFPRI forecasts a possible 50% fall in its wheat yield in 2050 (one-sixth of all the world's wheat grows on the north Indian plain). In the Middle East the institute predicts yield declines of 47% for maize and 30% for rice. As patterns of production shift, argues Jerry Nelson, the report's lead author, it becomes all the more important to liberalise farm trade, so that farming keeps pace with changing comparative advantage. But overall, he argues, the yield declines are so great that only another round of technological changea new Green Revolutionwould be enough to offset them. In principle, such a thing looks possible: the technology to double or triple many crop yields exists in laboratories. The problem is to get it into the fields. To that end, last week's G20 meeting in Pittsburgh promised to put more taxpayer money into farm research and other help for agriculture. The ups and downs of diplomacy used to be compared to the cycle of the seasons. But as poor smallholders are finding out, the seasons are not what they were. * Climate change: impact on agriculture and the costs of adaptation. By Gerald Nelson and others. IFPRI.