Here is a mind-boggling estimate from Jason Clay of the World Wildlife Fund: We will need to produce 2.5 times as much food in the next 90 years as we have in all of the last 8,000 years combined. Or, more than a factor of three increase in annual production in this century alone. This is a direct result of the world’s population exceeding 10 billion by end of this century, accompanied by a doubling of per-capita consumption.
Food production already occupies 58 percent of Earth's habitable land and accounts for 67 percent of fresh water consumption. Climate change goes hand in hand with this. According to the FAO, agriculture directly contributes 13.5 percent of global GHG emissions. With the additional impacts of land-use changes, food processing and the rest of the value chain, the provision of food likely exceeds a quarter of all GHG emissions.
Given the limited availability of additional land, water and energy – coupled with the need to first cap and then reduce GHG emissions – much of the daunting challenge of sustainably feeding the world’s population will have to be met through dramatic increases in efficiencies. We can talk about increasing two very different types of efficiencies. One is the efficiency of food production and supply. The other is the efficiency of food consumption.
The supply-side efficiencies depend on better technologies and practices on multiple fronts, including: increasing the productivity of land and input use by a factor of 2-4; cutting waste and spoilage in processing, storage and distribution (especially in developing countries); and leveraging carbon storage in soils (potentially as much as 6 Gt of CO2e/year) to reduce net agricultural emissions.
This will involve many different technologies, but chances are that some of the ideas commonly associated with food sustainability – such as organic production, locally produced food, grass-fed or free-range animals, and the like – are unlikely to make a big difference. Here is why:
--Life-cycle assessments of organic production – including our recent study – point to a number of common inefficiencies such as lower yields and higher on-farm energy use. Soil carbon sequestration from the application of manure and other organic inputs remains the one clear advantage in the first few decades of transition after converting conventional croplands or other degraded lands to organic production.
--Transport generally contributes less than 10 percent to life-cycle GHG emissions and is therefore not a high leverage point. Moreover, local food production often suffers from inefficient distribution compared to the highly streamlined logistics of long-distance goods transport. The one significant exception is air-freighted food.
--Recent research has shown that grass-fed beef raised in established pastures in the Upper Midwest produces about 30 percent higher GHG emissions than comparable feedlot-finished beef, primarily because of the lower-quality diet. On pastures transitioning to management-intensive grazing, the pastured beef produces 15 percent lower net emissions than feedlot beef during the transition period due to increases in soil carbon – still far short of the efficiencies needed for beef production.
--Our analysis of a few instances of free-range poultry and swine production suggest that the feed input requirements are similar to confined systems (the animals get little or no nutrition from foraging), but the efficiency of food production is lower. (This is only an efficiency argument and not meant to condone unethical treatment of animals.)
If we move away from rigid, binary classifications of food production – such as organic and non-organic – we might agree that the solution space is more continuous and larger than previously thought and includes many more possibilities. For example, it should be feasible to design highly optimized hybrid farming systems that combine management practices to increase soil organic carbon stocks with ultra-efficient conventional systems to produce high yields at a lower footprint.
WWF’s Jason Clay lists crop genetics – including GMOs – as the most important supply-side strategy to freeze the footprint of food. Genetic engineering in the food sector is the big elephant in the room, not unlike nuclear power in the energy sector: The upside can be wonderful and the downside can be frightening. But we are already using genetic manipulation widely and will likely need to focus on safe and strategic ways to integrate it into optimized food systems to improve the yields of high-volume crops.
The consumption-side efficiencies involve finding ways to meet nutritional, taste and aesthetic needs at the lowest possible level of resource use and cost to consumers. We know that food consumption is going to increase dramatically in the coming decades and some of the key drivers – such as the general population trajectory, higher per-capita consumption and the need to provide basic food security to everyone – are already well established. However, there are areas where consumption could be made much more efficient and thereby relieve some of the immense pressure on the supply side.
The first of these areas is food waste. A recent report estimated that food waste in the UK is large enough to account for three percent of domestic GHG emissions. Our analysis of US food waste came to a similar conclusion: about a third of the food reaching the retail stage is ultimately wasted, amounting to at least two percent of national GHG emissions on a life cycle basis.
While most of the food waste in developing countries occurs in storage and distribution (which is a supply-side problem), the biggest share of waste in developed countries occurs at the consumer level both at home and away from home. There is clearly a need for smart food management solutions that consumers, retailers and the food service industry can all use to get the most out of the food that we are already producing. Any significant reduction in food waste will free up resources that can be used to feed more people elsewhere.
A second consumption-side issue is how we obtain our nutrition. Our analysis of USDA’s new MyPlate framework for building healthier diets shows that GHG emissions (and indirectly agricultural resource uses) are dominated by protein consumption. In a typical diet conforming to the MyPlate recommended amounts for each food group, the largest share of emissions comes from animal proteins. We can cut per-capita agricultural emissions and resource use significantly by getting most of our proteins from plant sources. National food service companies such as Bon Appétit and Sodexo, as well as mainstream food writers such as Mark Bittman of the New York Times, are now advocating plant-based diets augmented with small amounts of meat.
As developing countries become more affluent, consumers there will increasingly need to contend with the same consumption-side issues. While much more food needs to be produced in the coming decades using both existing and yet-to-be-developed technologies, we also need to make the most of every pound of food that we already know how to produce.
We do have an existing model for a combined supply-side and consumption-side solution in the energy sector: vigorously promote energy efficiency and conservation while expanding the energy supply through new technologies and investments. Why not take the same approach when it comes to food?
Kumar Venkat is president and chief technologist at CleanMetrics Corp., a provider of analytical solutions for the sustainable economy.