Nonlinear Temperature Effects Indicate Severe Damages to US Crop Yields Under Climate Change

The United States produces 41% of the world’s corn and 38% of the world’s soybeans. These crops are two of the four largest sources of caloric energy produced and are therefore critical for world food supply. With the evidence that greenhouse gases are warming the world’s climate it is pertinent that we understand the effect of temperature on crop yields. This study links the relationship of weather and crop yield of soybeans, corn, and another warm weather crop, cotton, the crops with the largest production values in the United States. The data that are used in the study are composed of a new fine-scale model of weather outcomes merged with a large panel of crop yields from most U.S. counties in the time span of 1950–2005. The new weather data include length of time each crop is exposed to each one-degree Celsius temperature interval in one day. Then it is summed across all the days of the growing season within each county. The result of this study is that high temperatures are much more damaging than previously thought and that yields of all three crops are likely to decline greatly if warming predictions are accurate. Hannah Carr
Schlenker, W., Roberts, M., 2009. Nonlinear temperature effects indicate severe damages to U.S. crop yields under climate change. PNAS 106, 15594–15598.

The study shows that yield growth increases gradually with temperature up to 29–32 °C, depending of the crop, and then decreases sharply for all three crops. For the corn, the critical threshold temperature is 29 °C, for the soybeans it is 30 °C and for cotton it is 32 °C. With the current growing regions fixed, average yields are predicted to decrease by 30–46% before the end of the century, with the slowest warming scenario and decrease 63–82% with the most rapid warming scenario. The driving force behind these large and significant predicted impacts is the projected increase in frequency of extremely warm temperatures.

This study of crop yields uses a system of models in which each model is estimated 1,000 times, randomly choosing 48 years of the 56 year history. The models are compared with three specifications of temperature effects (step function, polynomial, and piecewise linear) along with three alternative specifications: 1) a model with average temperatures for each of four months 2) an approximation of growing degree days based on monthly average temperatures and 3) a measure of growing-degree days, calculated by using daily mean temperatures. To demonstrate the study throughout the country, it divided the United States into three regions: the northern cooler states, the southern warmer states and the middle states. I these regions the study used corn as it is a main crop across the entire country. It was found that there is a nonlinear relationship between yield and temperature throughout the country, and that greater precipitation will partially mitigate the damages from extreme temperature increases. The study also proves that estimated climate change impacts are not sensitive to the specific growing season and consistent with time separability. The findings of the study are notable for the consistency of the estimated nonlinear temperature effects across time, location, crops, and the many sources of variation in temperature and precipitation considered. 

Response to Schlenker and Roberts’ Study of US Crop Yields under Climate Change

In response to Schlenker and Roberts’ article Nonlinear temperature effects indicate severe damages to U.S. crop yields under climate change, which indicated that temperature increases will severely damage US crop yields in the next century Meerburg et al. (2009) challenged the paper. Meerburg et al’s article Do nonlinear temperature effects indicate severe damages to US crop yields under climate change states that in fact the Schlenker and Roberts article is erroneous, and that it was a pessimistic view of the issue. Hannah Carr
Meerburg, G., Verhagen, A., Jongschaap, R., Franke, Schaap, B., Dueck, T., van der Werf, A. 2009. Do nonlinear temperature effects indicate severe damages to US crop yields under climate change? PNAS 106, 120–120.

The study by Meerburg et al. states that high temperatures have different effects on plants at different developmental stages and are not always problematic. They used Brazil as an example of temperature having little to no effect on crop yield. The article states that Brazilian farmers successfully increased the productivity of soybeans, maize, and cotton in the last decade despite the fact that the cumulative days of exposure to temperatures above the threshold values is far greater than in the US. The state of Mato Grosso, Brazil exceeds 35 °C for 118 days per year, 75 of those days being in the average soybean-growing season and the production of soybeans was approximately 3.1 tons ha-1 yr -1 in 2008, exceeding that of the 2008 US crop yield. Similarly the cotton yield of Brazil in 2006/2007 was 1.4 ton ha-1 yr-1 compared to the US’s .9 ha-1 yr-1. With these two examples the authors concluded that despite relatively long periods of exposure to high temperatures, Brazilian farmers have managed to boost crop productivity.
With the challenge at hand, Schlenker and Roberts reply, stating that Meerburg et al’s. information is misleading because some areas of the United States have much greater exposure to extreme heat than Mato Grasso and vice versa. The authors contrast Mato Grosso to Illinois, which is one of the most productive states in the United States. The average daily summer temperatures of Mato Grosso are in fact higher than in Illinois, however after a satellite scan of the region it reveals that there are in fact no soybeans growing in the hotter area. If the maximum temperatures are averaged over the region that soybeans are actually grown in the average exposure to degree-days less than 30 °C in Mato Grosso, similar to the southern half of Illinois. Therefore, the authors state that there is no contradiction between crop yields in Mato Grosso and the earlier findings in the United States.

Rice Yields Decline With Higher Night Tempera-tures From Global Warming

World rice production must increase by 1% annually in order to meet the growing food demand that results from population growth and economic development. Peng et al. (2004) report on the impact of global warming on rice yields at the International Rice Research Institute (IRRI) in Los Baños, Laguna, Philippines. The study was done using simulation models of rice crops. The purpose of the study was to observe the direct effects of climate change on crop production. From 1979 through 2003 weather data were analyzed to examine temperature trends and the relationship between rice yield and temperature. Hannah Carr
Peng, S., Huang, J, Sheehy, J., Laza, R., Visperas, R., Zhong, X., Centeno, G., Khush, G., Cassman, K., 2004. Rice yields decline with higher night temperature from global warming. The National Academy of Sciences of the United States of America 101, 9971–9975.

The simulation was carried out on the IRRI research farm in a 10.5 by 9.5 m plot surrounded by irrigated rice. At this site measurements of dry- and wet-bulb temperatures, and minimum and maximum air temperature and radiation were taken daily. In contrast, the simulation models were developed using rice seedlings raised in trays in a greenhouse. Different amounts of P, K, and Zn were applied to the seedlings in the dry and wet seasons.
The field study showed that for each °C increase, grain yield will decrease by at least 10%. The study confirms predictions from simulation studies of sustainable yield reductions caused by higher mean daily temperature but yield reductions in the simulations tended to be smaller. In the trays, increasing night temperature from 21 °C to 29 °C at a constant day temperature of 29 °C reduced the production of the rice plants by approximately 20%. The magnitude of grain-yield reduction from an increase in mean daily temperature is similar to the 17% decline per °C in the United States study of the trends in climate and its effect of maize and soybean yields from 1982 to 1998. The report displays direct evidence of decreased rice yields from increased night temperatures. However, the results of the study also highlight the need for a greater understanding of the effects of night temperature on physiological processes in crop growth and yield development.