The food distribution gaps projected for 2012 (Section 220.127.116.11) implicitly include the effects of climate change because they were based in part on historical trends. The long-term impacts of climate were isolated from other factors by estimating total percent changes in growing season for the 2002-2012 period, quantifying the relationship between growing season lengths and crop yields, removing the climate-induced changes in yield from the base yield projections, and re-estimating the food distribution gaps in 2012.
Total percent changes in cropland growing season for the 2002-2012 period were estimated by simply extrapolating data from the estimated percent changes in growing season during the 1902-1997 period. The changes were in the same direction, but they were smaller in magnitude than the 1902-1997 impacts depicted in Figure 4.4. Length of growing season is an important component of agricultural productivity (FAO, 1996). Along with other factors, it determines what crops can be grown in a particular area. It also determines the number of crops that can be grown sequentially during the year.
In general, longer growing seasons mean greater agricultural productivity. Darwin et al. (1995), for example, estimated that in 1990 the average amount of crop produced on all cropland with short growing seasons of 100 days or less was 1.38 metric tons (mt) per hectare, while the average amount of crop produced on all cropland with long growing seasons of 301 days or more was 4.80 mt/ha. This is equivalent to 0.9% per day of additional crop produced (0.9% = 100 x [(4.80 mt/ha - 1.38 mt/ha)/1.38 mt/ha/(333 days - 50 days]). In this analysis, the link between growing seasons and yields is quantified by assuming that a 1% change in growing season is accompanied by a 1% change in yield. The climate-induced changes in yield are then removed from the base yield projections and food distribution gaps for 2012 are re-estimated. The difference between the original and revised estimates is attributable to climate change.
Results of the analysis are depicted in Figure 4.7. Estimated impacts from SOI- and CTI-based models were similar, as shown by the 0.975 correlation between them. Only in India are impacts predicted to differ. Except for Uzbekistan, the 27 countries that are food secure in 2012 are not affected by climate change. In Uzbekistan, a distribution gap is projected under the no-climate-change scenarios. This indicates that without a climate-induced increase in an average growing season, Uzbekistan would be food insecure in 2012. Distribution gaps in another five to seven food-insecure countries are also unaffected by climate change.
In 22 to 25 of the remaining countries, however, climate change is estimated to increase the distribution gap. In some countries, climate-induced decreases in average growing season are estimated to contribute to reductions in food security of greater than 5%. These include Cameroon, Mali, Nigeria, Niger, Togo, Chad, Senegal, Sudan, Guinea-Bissau, and Mauritania in Sub-Saharan Africa; Bangladesh in Asia; Bolivia and Honduras in Latin America; and Armenia among the former Soviet republics. In the 13 to 14 countries where climate-induced increases in average growing season are estimated to improve food security, the impacts tend to be more modest, or less than 5%. Uzbekistan and India are the exceptions, but the impacts in India are uncertain in that the SOI-and CTI-based models generate opposite impacts. The relative contribution of climate change to the total food distribution gap for the low-income countries evaluated in this analysis is less than 0.1%.
Was this article helpful?