The global warming challenge for wheat

David Lobell is an associate professor in environmental earth system science and deputy director of the Center on Food Security and the Environment at Stanford University.

Scott Chapman is a principal research scientist with the Commonwealth Scientific and Industrial Research Organisation (CSIRO).

Wheat likes it cool. More than any other major food staple, wheat yields suffer when weather gets warm. Although wheat has been adapted to grow in almost every country of the world through 10,000 years of farmer selection and a century of breeding, it is still most productive in the cooler places and seasons. Given this reality, it is not surprising that most projections of climate change impacts indicate decreases in wheat production, with an average of roughly 5 percent yield loss expected for each 1°C of warming.

Indeed, many studies indicate that in recent decades global yield productivity has already been hit by warmer average temperatures.

But how important will global warming really be in shaping future wheat supply? The answer is hard to pin down, partly because it will depend on where and how quickly the world heats up.

The expected average rate of warming over current wheat areas for the next few decades is a little less than 0.5 °C per decade, which implies a negative yield impact of about 2 percent per decade.

Compared to the anticipated yield growth needed to keep pace with demand for wheat – about 15 percent per decade – this is a significant but modest addition to an already hard task.

But there is no guarantee that the expected rate of warming will happen, and models suggest that warming of as much as 1.0 °C per decade is plausible over the next couple of decades. Therefore, it is reasonable to view global warming as a major risk to future wheat supply.

What can the wheat community do to reduce these risks to the wheat supply, or more specifically, what might be done differently than business-as-usual wheat breeding? We have three suggestions.

First, in a shifting environment, it is expected that crops will face different conditions in farmers’ fields than they did in breeding trials. Physiological models that can skillfully predict crop growth then become an indispensable tool, because they can help breeders better anticipate what traits will prove useful even if those traits don’t confer yield advantages in their trials. This is an extension of a long-standing challenge of understanding gene-by-environment interactions, but requires new skill in modeling mechanisms behind crop responses to heat.

Second, these desired traits should be more reliably and quickly incorporated into elite wheat germplasm. This work will require a combination of methods to rapidly identify desirable traits among a large population (for example, rapid phenotyping) and methods (such as marker and genomic selection) to introduce and recombine new genetic variability when needed (for example, by using wild relatives).

A challenge here is that potentially adaptive traits, such as changes in flowering time, increased water-use efficiency or expression of “sunscreens” (surface waxes), might exist in current germplasm, but be genetically linked to less desirable traits. An ability to generate and test many combinations of different traits is more important, and harder, than simply having access to extreme values of a particular trait.

Third, agronomy will play a key role in helping the world adapt to climate change. For example, new planting methods that allow earlier sowing can help to escape the end-of-season heat and targeted use of mulches and irrigation can help to lower canopy temperatures. Widespread testing of these techniques, ideally in combination with the testing of various genotypes, would help to ensure that promising approaches are more quickly identified and scaled up.

In a world without climate change, these issues are still relevant. But they become crucial in a world where the risks of large heat waves rise each year.

We see global warming as an important strategic issue for international groups like CIMMYT, because: (i) warming is expected to be fast enough to significantly slow global yield growth; (ii) it is difficult to predict exactly which countries will see the most severe heat waves in the next couple of decades; and (iii) it is likely that some (if not all) countries will need germplasm that is currently grown elsewhere to adapt. Thus, the global threat is serious, but individual countries have limited incentives to devote significant effort to adaptation (because they may be spared the worst of it) and limited capacity to achieve success on their own (because of the need for imported germplasm).

It will take global institutions to successfully adapt to global warming.