Trends in maize, rice, and wheat yields for 188 nations over the past 40 years: a prevalence of linear growth

To prevent or prepare for future food shortages an understanding of the likely magnitude and distribution of future cereal yields is required. To this end, predictions of cereal yields have commonly been made, using various assumptions. However, the employed assumptions, namely, that yields tend to follow a given trend over time, have not been extensively tested. This study presents a test of the applicability of two general models to time series of maize (Zea mays L.), rice (Oryza sativa L.), and wheat (Triticum aestivum L.) yields for 188 nations to characterize past yield trends, to assess the relative importance of various trends on a global scale, and lastly, to determine what factors might be responsible for the presence of slowing yield growth and yield decline in some nations. Results showed that linear growth in yields has been the most common trend over time, occurring in more than half of all nation-crop data sets, and that growth significantly greater than 33.1 kg ha(-1) yr(-1) (the rate at which global cereal yields must grow to have the current per-capita production in 2050) constituted 20% of the data sets and was the most important trend in terms of global area harvested, production, and population. A trend of slowing yield growth was present in roughly one-sixth of the data sets, and the nations that this subset comprised made a small contribution to global area harvested, production, and population (less than 10%). Nation-crop data sets that showed yield growth greater than 33.1 kg ha(-1) yr(-1) had much greater yields than those that showed slowing yield growth, demonstrating that yield growth is not being limited by general physiological constraints to crop productivity. The results of a logistic regression procedure showed that the relative frequency of slowing yield growth and yield decline was negatively correlated to per-capita gross domestic product (GDP) for maize and wheat, and to growth in fertilizer rate for maize. In addition to GDP, latitude was negatively correlated with the relative frequency of yield decline. There were no significant predictors for rice. These results suggest that both economic and biophysical factors have played a role in limiting cereal yield growth. (C) 2003 Elsevier Science B.V. All rights reserved.