With the next two decades set to see a stronger increase in demand for food than the rest of the 21st century, declining harvests would cause some serious problems. Right now crop yields are growing, but could climate trends cause them to fall by a tenth, say, over the next 10-20 years?
That’s the question David Lobell from Stanford University in California and Claudia Tebaldi from the US National Center for Atmospheric Research in Boulder, Colorado have tried to answer. They find that if the world wasn’t warming, the chance of yields decreasing by a tenth over the next 20 years would be less than 1-in-200. However, climate change has made shrinking yields more likely, shortening the odds to a 1-in-10 chance for maize and a 1-in-20 chance for wheat.
“It was surprising to see how likely it is nowadays for climate trends to significantly cut into yield progress,” David told me. “It is still more likely than not that climate will be a slight drag on progress instead of a major factor. But we can’t rule out a major slowdown, and that means we should probably think through that type of scenario to figure out how to prepare for it.”
Such near-future climate forecasts are unusual, David underlined. “Longer periods allow the signal of climate change to become clearer compared to natural variability,” he explained. “But it may simply be that most of the initial questions about climate change were about the long timescales, to decide about questions related to energy choices and emissions. Now, a lot of questions are related about how to properly adapt to the changes happening now.”
What will happen to crops is central to David’s interests as associate director of Stanford’s Center on Food Security and the Environment. “I often get asked by governments or the private sector if climate change will threaten food supply in the next couple of decades, as if it’s a simple yes or no answer,” the scientist revealed. “This was especially true of a committee I recently served on focused on social stresses from climate change in the near-term. The truth is that over a 10 or 20 year period, it depends largely on how fast things warm, and we can’t predict that very precisely. So the best we can do is put odds on things.”
Untangling the roles of climate and better farming
In an Environmental Research Letters paper published last week, whose title expresses a desire to stop us ‘Getting caught with our plants down’, David and Claudia estimate those odds. To do that, they first had to work out how rising temperatures and rainfall affect wheat and maize yields. That’s not simple to isolate, because harvests are also affected by other factors, such as improving farming methods. Lacking data for these influences, they used an approach known as ‘first differences’ used in economics when information is missing. “The sensitivities that we infer for these crops – about 6-7% yield loss per degree of warming – agree well with estimates from other approaches, including experiments or using crop simulation models,” David said.
But because Claudia and David sought to explore uncertainty, they fed 500 random values around these sensitivities into two sets of the most modern and rigorous climate models available. “It’s critical for this type of question to explore a full range of possible climate scenarios, not just a handful, since we are trying to look at outcomes that are not very likely,” David said. “Using two ensembles – one based on running lots of models and one based on running one model for lots of different initial conditions – we were able to get a better sense of the chance of these outcomes.”
Using these figures and models, the American scientists then compared scenarios where greenhouse emissions and warming continue against simulations with no emissions, where the world doesn’t warm. Their results show that because of where maize is farmed, it’s slightly more likely to experience an average 1°C temperature rise at a critical time of year than wheat.
Not knowingly over-certain
The simulations also suggest that there’s a 1-in-10 to 1-in-5 chance of a cooling trend over the next decade, though that shrinks to around a 1-in-20 chance by the end of the second decade. Average yield impacts are similar for maize and wheat – both falling by a fortieth after one decade and a twentieth after two. However there’s a much wider spread of possibilities for maize, which means the chances of much worse harvests are higher than for wheat.
David expects the details of the study will draw criticism from other scientists who think there are better ways to estimate these odds. “That’s all for the better if more people start focusing on the issue of near-term risks and getting better numbers,” he said. “For people working on issues related to food security in governments, NGOs, and the private sector, hopefully the general point of the paper will help to motivate more thinking about how to ensure continued progress in crop yields. This paper won’t be very useful for people who want to know what will definitely happen. But there’s a whole community of people used to dealing with risk and planning for contingencies.”
Claudia adds that their paper is part of a greater effort among climate scientists to account for natural variability and other sources of uncertainties, like those in climate models. She points to a paper published in the journal Nature last year examining the forthcoming effects of climate change on ecological systems that didn’t do this. Last week a different group of researchers warned the original authors hadn’t been through the uncertainties in enough detail, and therefore presented forecasts as more precise than they really were.
“Our study took advantage of large ensembles of simulations in order to avoid just that –misrepresenting results to be too precise or too certain,” she explained. “We’re trying to get at the odds and risks, accounting for the fact that the climate system, even when undisturbed by greenhouse gas emissions, shows some odds of the events we are concerned about. We want to characterise the enhanced risks as best we can, when we see them in our data.”
Lobell, D., & Tebaldi, C. (2014). Getting caught with our plants down: the risks of a global crop yield slowdown from climate trends in the next two decades Environmental Research Letters, 9 (7) DOI: 10.1088/1748-9326/9/7/074003
Mora, C., Frazier, A., Longman, R., Dacks, R., Walton, M., Tong, E., Sanchez, J., Kaiser, L., Stender, Y., Anderson, J., Ambrosino, C., Fernandez-Silva, I., Giuseffi, L., & Giambelluca, T. (2013). The projected timing of climate departure from recent variability Nature, 502 (7470), 183-187 DOI: 10.1038/nature12540
Hawkins, E., Anderson, B., Diffenbaugh, N., Mahlstein, I., Betts, R., Hegerl, G., Joshi, M., Knutti, R., McNeall, D., Solomon, S., Sutton, R., Syktus, J., & Vecchi, G. (2014). Uncertainties in the timing of unprecedented climates Nature, 511 (7507) DOI: 10.1038/nature13523