The first stage of the scientists' research indicated that yields had historically increased as temperatures rose to an optimum value of 29°C for corn, 30°C for soybeans, and 32°C for cotton. At temperatures above these optimum values, however, crop yields declined, and they did so with slopes that were significantly steeper than the upward slopes that had preceded them. Then, in the second stage of their research, they found that "holding current growing regions fixed, area-weighted average yields are predicted to decrease by 30-46% before the end of the century under the slowest warming scenario and decrease by 63-82% under the most rapid warming scenario under the Hadley III model."

On the downside, this was not good news. But on the upside, it was much too bad to be true; and about six weeks later, the Proceedings published a letter by Meerburg et al. (2009) that provided a whole new perspective on the issue.

The seven Dutch scientists began their critique of Schlenker and Roberts' study by noting that yields of the crops in question will continue to increase in years to come, because of "the development and adoption of new technologies and improved farm management," citing in this regard, the results of Ewert et al. (2005), which indicate that continuing advances in technology have historically been the most important driver of productivity change, even outweighing the negative effects of detrimental climate change. And in further illustration of this phenomenon, they report that between 1961 and 2007, "average US corn yields increased by 240%, from 3.9 tons per hectare per year to 9.4 tons per hectare per year (FAO, 2009)," while noting that some researchers have predicted that "advances in agronomics, breeding, and biotechnology will lead to an average corn yield in the US of just over 20 tons per hectare per year in 2030," citing Duvick (2005).

Meerburg et al. also make note of the fact that farmers in Brazil successfully increased the productivity of soybeans, maize, and cotton during the last decade, despite the fact that the cumulative number of days of exposure to temperatures above the three crops' optimum values "is far greater than in the US." In the Brazilian state of Mato Grosso, for example, they say that "maximum average day temperature exceeds 35°C for 118 days per year, of which 75 days are in the average soybean-growing season." Nevertheless, they report that in 2008 average production of soybeans was about 3.1 tons per hectare per year in Mexico, while the average yield in the US was 2.8 tons per hectare per year. Similarly, they note that the mean cotton yield in Brazil in 2006/2007 was 1.4 tons per hectare per year, while in the US it was only 0.9 tons per hectare per year.

The seven scientists thus conclude that "temperatures higher than currently experienced in the US do not necessarily need to coincide with lower crop yields and that already existing technology and future advances (new varieties, optimized farm management, biotechnology, etc.) can overrule the negative effect of increasing temperatures on yield," as has in fact been observed to be the case in the historical crop yield data of the US.

A final problem with the analysis of Schlenker and Roberts (2009) is their acknowledged "inability to account for CO2 concentrations," the increasing levels of which, in their own words, "might spur plant growth and yields," such that "yield declines stemming from warmer temperatures therefore may be offset by CO2 fertilization," as has indeed been found to be the case by many of the studies we have discussed on our website, reviews of which are archived in our Subject Index under the several sub-headings of Growth Response to CO2 with Other Variables (Temperature).

In light of Schlenker and Roberts' stated admissions and the facts cited by Meerburg et al. -- which should have been known by the two US researchers, as well as the esteemed communicator of their paper to the Proceedings of the National Academy of Sciences and the appropriate editorial staff of the prestigious journal -- it is clear that their paper should never have been published, especially with a title that proclaims as fact that "nonlinear temperature effects indicate severe damages to U.S. crop yields under climate change."

This has got to be a prime example of "peer review" and unwarranted publication at their very worst!

Sherwood, Keith and Craig Idso

References
Duvick, D.N. 2005. The contribution of breeding to yield advances in maize (Zea mays L.). Advances in Agronomy 86: 83-145.

Ewert, F., Rounsevell, M.D.A., Reginster, I., Metzger, M.J. and Leemans, R. 2005. Future scenarios of European agricultural land use: I. Estimating changes in crop productivity. Agriculture, Ecosystems and Environment 107: 101-116.

FAO. 2009. FAOSTAT Database. United Nations Food and Agriculture Organization. Available at www.fao.org. Accessed 8 September 2009.

Meerburg, B.G., Verhagen, A., Jongschaap, R.E.E., Franke, A.C., Schaap, B.F., Dueck, T.A. and van der Werf, A. 2009. Do nonlinear temperature effects indicate severe damages to US crop yields under climate change? Proceedings of the National Academy of Sciences USA 106: 10.1073 pnas.0910618106.

Schlenker, W. and Roberts, M.J. 2009. Nonlinear temperature effects indicate severe damages to U.S. crop yields under climate change. Proceedings of the National Academy of Sciences USA 106: 15,594-15,598.