Blog
The high importance of low(er) expectations, energy crop yields edition
Commercial-scale cellulosic biofuels are a kind of holy grail for energy policy: a way to replace fossil fuels with low-carbon, low-impact renewable energy while avoiding problems inherent in the alternative approach of using food crops (like corn) to produce fuel. Energy crops don’t produce food. Rather, their leaves, stems, and stalks are burned for electricity or turned into biofuel. One of the attractions of these plants is that they grow faster than food crops currently used to make biofuels like corn ethanol.
But as my colleague Chris Malins and I point out in a paper just published in Biomass and Bioenergy [paywall], energy crops probably don’t grow as fast as much of the current research on them assumes that they will. And that, in turn, signals a problem for policy based on that research.
How fast energy crops grow matters for two reasons: [1] The faster a crop grows, the more money the farmer can make per field, so lower yielding crops would make cellulosic biofuel more expensive than expected. [2] There’s only so much unused land on the planet, and the rate at which we can grow biomass on this land limits the maximum amount of this renewable resource we could use for energy in the future.
This second issue has been a big question for intergovernmental organizations that want to map out our global energy future, like the International Energy Agency (for example, here [.pdf]) and the Intergovernmental Panel on Climate Change (here). Both the IEA and the IPCC predict that we could potentially meet a very large fraction of our future energy needs through energy crops. But these projections are based on modeling studies [paywall] that assume really high energy crop yields. For instance, the IEA assumes [.pdf] we can grow woody crops like willow or poplar at an average rate of 15 tons per hectare per year, which is substantially higher than typical U.S. corn yields of 10 t/ha/yr. After reviewing the literature on five types of energy crops, we found that a range of about 4–10 t/ha/yr can be expected for commercial scale willow and poplar, and 5–15 t/ha/yr for Eucalyptus grown in the tropics. So the average yield at a global scale would probably be around 8 t/ha/yr, about half that assumed by IEA.
Why does the IEA predict we can achieve high yields for these crops? It’s possibly because the literature on energy crop yields is easily misunderstood. Some peer-reviewed studies do report very high yields, for instance a whopping 51 t/ha/yr for Eucalyptus in one study. But for a number of reasons we shouldn’t take numbers like this at face value.
First off, the extremely high Eucalyptus yield was extrapolated from a minute plot of 5 trees, grown under ideal conditions in the mild climate of northern New Zealand. Cherry-picking the highest yield from a very small area of land does not tell you what yields will be over a commercial field of several acres with some pockets that have poorer soil or drainage than others. Additionally, small plot experiments typically have an open corridor between plots that provides extra light to plants growing on the edge. The smaller the plot, the more this edge effect can boost growth in a way that wouldn’t happen in one large continuous field.
Another problem is that these plots are often on prime agricultural land (like our plot in New Zealand, or a rich, well-draining soil in Iowa) and are carefully managed. In some cases weeds are hand picked and the crops are irrigated and fertilized. This isn’t the kind of biomass production we talk about delivering high carbon savings with no impact on food markets. In fact, the modeling studies described above generally assume that biomass crops will not compete with food production. Instead, they assume that only “marginal land” that is unsuitable for food crops will be used for energy production. But this type of land isn’t used for food production precisely because it is low yielding, with poor soil, low rainfall or cold temperatures. When one study [paywall] tried to grow Eucalyptus on very dry land in India without irrigation or fertilizer, the authors measured yields of 0.4 t/ha/yr, 100 times lower than the New Zealand study.
Lastly, researchers usually carefully hand pick every last twig and leaf when harvesting biomass in these small plot experiments. In commercial production, some biomass losses are unavoidable. The biomass can’t be stored fresh and so perennial grasses are typically left out to dry in the fields overwinter and harvested in the spring. Over these months, some biomass is inevitably lost. Combines (large machines that harvest the crop) cut the stem several inches to a foot above the ground, losing the base of the crop.
Many researchers quite reasonably expect yields of energy crops to improve in the future. After all, corn yields have more than doubled over the past 50 years. But can these same gains be realized in Miscanthus, switchgrass and Eucalyptus? There are several reasons to expect not. Much of the yield gains in crops like corn and wheat has come from increasing the ratio of grain to straw—that is, you get more corn but less leaves and stalk. This breeding technique won’t help if you need the entire plant. Energy crops generally have longer breeding periods than food crops. For example, you could grow a corn or soybean plant to reproductive age in just a few months, and this would let you selectively breed multiple generations within a year. But biomass crops like willow take several years to reach maturity, and this dramatically slows down the speed of research. Some energy crops have particular challenges. The high-yielding type of Miscanthus is a hybrid between two different grass species, and (sort of like a mule) it’s sterile and cannot produce offspring. Effectively, Miscanthus can’t be bred, and this severely limits the research options for increasing yield.
There is still promise for energy crops to deliver real environmental benefits and we should continue to direct research and investment towards cellulosic biofuels. Besides wastes and residues, these crops represent our best chance at producing clean energy with low impacts, and may even have direct environmental benefits like supporting biodiversity and sequestering carbon in soils. Lower than expected yields may make commercialization of energy crops a little more challenging, but that doesn’t mean it’s not a challenge worth taking on.