January 17, 2014
You hear it all the time in agricultural policy and research discussions: yields for the world’s major cereal crops will continue marching steadily upward. In fact, yields in many parts of the world have already plateaued and the relative rate of increase everywhere else is declining, says Ken Cassman, Robert B. Daugherty Water Food Institute Fellow and University of Nebraska-Lincoln agronomist.
In a study recently published in the journal Nature Communications, Cassman and UNL colleagues Patricio Grassini and Kent Eskridge suggest the challenge to feed the world is even more daunting than previously thought.
We spoke with Cassman about their findings and what it means for the future of global food security. Cassman leads an international research effort to create the Global Yield Gap and Water Productivity Atlas, an easily accessible web-based platform to estimate exploitable gaps in yield and water productivity of the world’s major food crops.
Can you highlight the findings in your recent study?
There’s been a lot published about predicting future food supply and demand, and these studies are very influential because they’re used to support policy analyses for agriculture at national and international levels. They’re used to look at research priorities. They’re used to make policies about land-use and even climate change and biodiversity. As we all know, expanding agriculture into areas like rain forests, wetlands and grasslands is really harmful. These are the most carbon rich and biodiverse natural ecosystems remaining on the planet.
So the question is: how good are these analyses, particularly regarding their projections for future crop yields? We found that the scientific rigor brought to bear on evaluating past trends in crop yields to inform the assumptions used in future scenarios was very poor, almost casual. Some assumed that the historical trends of the past would just continue into the future. Some assumed that it might accelerate. Some even used exponential growth rates. It seemed to us that there was a need to apply a strong statistically acceptable standard to the way in which you analyze historic trends. And, in doing that, try to make some conclusions about what to expect in the future and what it means for research prioritization, food supply and demand, land use and so forth.
The scientific foundation of the paper is to develop this very rigorous statistical procedure by which you can take the yield trends of any country and determine the best mathematical form to fit that trend. So you take out any subjectivity, guessing, or back of the napkin kind of projections. We did that for all of the major producing countries for the major cereals: rice, maize and wheat.
The results were really quite striking.
First, there is no evidence of exponential yield gain. It’s surprising how many studies assume overly optimistic exponential gains, very influential studies published in some of the best journals and used to inform policy and research priority decisions.
Second, the linear model fits all of them very well, but sometimes there’s a break. In most cases that break either slows down or stops. We call that a yield plateau. So the second surprising thing is that in 31 percent of all maize, rice and wheat production currently in the world there is a statistically significant plateau or an abrupt decrease in the rate of yield gains in those cereals.
Are you saying that we can no longer assume steady rates of gains will continue into the future?
Exactly. In fact, we can take it a step further and say if you really want to be robust in your predictions, you’ve got to specify a yield ceiling. In other words, projections should allow yield rise, perhaps linearly, but at some point they reach a ceiling where it’s not likely that they’re going to increase further. We suggest there are actually biophysical limitations to yield depending on the amount of sunlight and water and the length of the growing season. You can actually estimate that for each country, if you wish. The Global Yield Gap Atlas project (LINK), funded in part by the Bill and Melinda Gates Foundation and by DWFI, is trying to do that for major crop producing countries worldwide.
Which areas are plateauing? Does that include U.S. agriculture, for example?
It sure does. We give an example for rice in California where there have been studies that look at future production using linear rates and even exponential rates and, yes, we see yields that have actually reached a plateau. That’s how far off you can be. It also looks like irrigated maize in the U.S. is near a plateau. We also see it for rice and maize in China and for wheat in India. It’s not always a plateau; sometimes it’s an abrupt shift to a slower rate of growth.
So these are areas where they’re making decisions based on linear or even exponential gains and yields are actually plateauing?
Yes, in nearly all cases. Isn’t that remarkable?
What does this mean for the future of global food security?
There are parts of the world where yields are very low, and it’s going to be a long time before they reach levels close to these biophysical limits. So those areas are clearly going to be very important for global food supply. These are places like sub-Saharan Africa. There has to be a lot more effort to ensure that marked productivity gains are made there. Part of the answer is research and development, but a significant part is using existing technologies. The chief limitations for so long have been lack of access to inputs, lack of markets, corruption, lack of infrastructure to move products and lack of ways to get information about appropriate technologies to farmers. So clearly that’s a major opportunity.
I think it’s just as important for more research and development efforts in developed countries. The main point is we really underestimated the magnitude of the challenge for developed countries like the United States. And yet there’s no way we can achieve food security, if yields don’t continue to increase in developed countries as well. Other areas, even if they massively increase production, are unlikely to stay ahead of population growth and rising incomes. What our paper suggests is that it’s not going to be as easy as people think and it’s going to take a lot of resources because, as you said, everyone assumes that linear rates will continue or, worse, are going to accelerate. We haven’t recognized that rates are going to slow unless we find new approaches for sustaining yields.
What does all of this mean for the water resources needed to grow food?
The same type of analysis can be used to estimate water productivity. With the same data used in the Yield Gap Atlas to estimate yield ceilings, you can estimate existing and potential water productivity. If water is your major limitation, you can look at those areas that are not using the available water efficiently and focus on them to close yield gaps. There’s a large gap between rain fed and irrigated yield potentials, but there are still a few remaining areas where it’s possible to expand irrigation. We need to identify those, but it has to be sustainable use of the water supply.
As you know, there is debate about the environmental consequences of increasing inputs, such as pesticides and water. How do your findings contribute to that debate?
Not directly, but intensification of agriculture is always a concern. The fact is that feeding humanity without expanding agriculture at the expense of the last remaining natural ecosystems mean you’ve got to get massively higher yields on the same land area. It’s not possible to do that without inputs or pest control. So the question is not whether you should use those inputs or not, but rather how you use those inputs without the negative environmental consequences.
I would put it in a positive way: if we focus on that as a goal, as a scientist, I have no concern that it can’t be done. But your research has to explicitly address the challenge of doubling yields on existing farmland while at the same time showing that you can do it with less loss of nutrients, less water pollution, less global warming intensity, without negative impacts on wildlife, without pesticides that damage soils. Otherwise, you’re not studying a system that’s going to be useful in the future.
Ken Cassman is the Robert B. Daugherty Professor of Agronomy at the University of Nebraska-Lincoln and a Robert B. Daugherty Water Food Institute Fellow.
Tags: food production, Global, research, yield gap