Current LCIA methods, for example, are not able to properly evaluate potential adverse effects of Bt toxin on populations of non-target species and elevated risk of species invasiveness through genetic modifications . In addition, it should be noted that the trend of decreasing ecotoxicity impact is unlikely to continue for cotton and corn. Due to the dominant use of HR and Bt crops, pests and weeds have evolved to be increasingly resistant . As a result, farmers may need to resort to earlier pest control practices that rely more on conventional pesticides, hence increasing crops’ freshwater ecotoxicity impact. Nevertheless, the dynamics of pest management, and associated ecological impacts, further corroborates the importance of understanding the dynamics of agricultural systems. For many of the impact categories studied, the environmental impacts of US corn and cotton on average are roughly comparable on a per hectare basis, while cotton consumes more water and generates much higher freshwater ecotoxicity impact. However, the average results, mainly reflecting corn produced in the Midwest and cotton produced in the South, are inadequate to capture the likely environmental consequences of corn expansion into cotton, which has taken place in cotton-growing states in the South. The state-level results show that a land use shift from cotton to corn relieves freshwater ecotoxicity but may aggregate many other regional environmental impacts. Due to the limitation of data, a definitive conclusion may not be drawn for other Southern states where the cotton-to-corn land use change has also occurred. But our finding of tradeoffs based on the three states is probably generalizable for these other states considering that cropland there is generally less suitable for corn production than in the Midwest. Taking into account marginal yield and technological advances, the CPT for converting the CRP grassland for corn ethanol production in early 2000s, when the ethanol industry begun to grow, 30 planter pot ranges from 15 years for highly productive grassland with average corn yield to 56 years for infertile grassland with only 50% of average corn yield.
Considering the diminishing climate effect of later GHG emissions within a 100-year time frame, the CPT estimates would increase to 17 to 88 years. Understandably, the shorter the payback time, the less strongly it would be affected by the consideration of emission timing.In the no technological advances scenario, most of the grassland would not produce any climate benefit within a 100-year time frame. Even for the highly productive grassland, it would take up to 46 years before the system could start generate carbon savings. Last, because the technology and productivity of the corn ethanol system changes over time, the timing of land conversion also plays a part in the CPT estimation. If land conversion took place in 2010, the CPT estimates would be 13 to 65 years, as opposed to 17 to 88 years for land conversion occurring in early 2000s. The environmental impacts per hectare crop harvested for most of categories studied were relatively stable in the past decade. This is because these impact categories are dominated by the direct and indirect emissions of nutrient, particularly nitrogen fertilizers, and the amount of nutrient inputs did not change much in the past decade. In contrast, the freshwater ecotoxicity impact per ha corn harvested declined by around 50% from 2001 to 2010 and per ha cotton produced declined by 60% from 2000 to 2007. These downward trends are due in large part to the increasing adoption of genetically modified organisms , which have resulted in reduced use of insecticides and replacement of some conventional herbicides with more benign ones, particularly glyphosate and compounds. Soybean production in the USA has also adopted GMOs widely, and this should have also led to a decline in soybean’s freshwater ecotoxicity as with corn and cotton. But because of the invasion of soybean aphid, a native of Asia, which have resulted in a substantial increase in insecticides use, the freshwater ecotoxicity impact per hectare soybean harvested increased by a factor of 4 from 2002 to 2012. In the meantime, on-farm irrigation water use per ha soybean harvested increased by about 50%.
This increase is due probably to the expansion of soybean into marginal land where intensive irrigation is needed. Implications of the above findings have been extensively discussed in individual chapters. Discussed below is the implication of considering marginal yield in the case of direct land use change for studies of indirect land use change and consequential LCA modelling. Some of the points, such as the importance of additional corn and carbon, have been somewhat touched upon in , but the discussion here is more detailed and from a methodological point of view. Early LCA estimates differed with respect to whether corn ethanol offers carbon benefits in displacing gasoline . Notably, the findings of the Cornell Professor David Pimentel were all negative , leading him to strongly oppose the use of corn ethanol . But subsequent LCA studies, with updated data and ethanol coproducts correctly accounted for, seemed to converge on that corn ethanol has a moderately smaller carbon footprint than gasoline, thus contributes to climate goals . However, a core factor was neglected in all these LCA studies, that is, land use change . The reason land use change did not come into play in these LCA studies is that they were basically a portrayal of exiting corn ethanol with corn grown on long-standing cornfield. But with increasing ethanol demand driven by federal policies like the renewable fuel standard aimed partly at mitigating climate change , what mattered was not the carbon footprint of existing corn ethanol but of additional corn ethanol. The key issue then became the supply of additional corn. Yield increase through intensification could produce more corn in the long run, but was hardly enough, and too uncertain, to meet annual ethanol expansion. The pressure was on land resources . Higher corn prices between 2005 and 2008 were driving farmers to bring new cornfield into production by converting natural habitats or to reallocate existing cropland to growing more corn . Either way, however, has dire carbon consequences that run counter to the initial climate goal of the federal policies.
Direct conversion of forest or grassland to grow corn for ethanol production would release a substantial amount of carbon stored in soil and plant biomass, creating a “carbon debt” that may take dozens of years to be repaid by carbon savings from substituting corn ethanol for gasoline . Similarly, reallocation of existing cropland to growing more corn could generate similar nets effects through market mediated mechanisms . For example, if the extra corn came at the expense of reduced soybean production, this could drive up global soybean prices and led farmers across the world to produce more soybeans by converting forest and grassland, resulting in loss of large amounts of carbon as well. In hindsight, that the majority of LCA studies failed to take account of land use change has a lot to do with the methodology they took, namely, attributional LCA . In these studies, corn ethanol’s carbon footprint was quantified in the simple accounting manner. They first estimated carbon emissions at different life-cycle stages based on existing, average corn farming practices and ethanol conversion technologies, and then summed them up and compared the total against the carbon footprint of gasoline. If they found that corn ethanol has a lower carbon footprint,plastic growers pots they would conclude that corn ethanol offers carbon benefits in displacing gasoline. Underneath the conclusion was the implicit assumption that the finding based on existing, average technologies would hold true for any amounts of additional corn ethanol. As argued above, however, the assumption is invalid. Because of land constraints, carbon emissions associated with additional corn ethanol would be much different from that associated with existing corn ethanol based on corn from long-standing cornfield . And it is the additional corn ethanol and associated carbon emissions that ultimately matter from both a policy perspective and in terms of reducing greenhouse gas emissions. In a word, consequential LCA looking into changes and effects is more relevant and better suited for addressing policy questions with potentially large economic and environmental consequences . But it should be noted that which specific methods to use for consequential modelling needs further research . The core to consequential modelling is the consideration of marginal changes, or processes actually to be affected by decisions at hand . In the case of dLUC, marginal changes include land conversion, additional corn production on the converted land, and additional ethanol produced and used. Particularly, the additional corn grown on the converted land sequesters additional carbon from the atmosphere. Without the additional carbon uptake, corn ethanol’s carbon benefits would not be possible as rightly pointed out by Searchinger . In short, it is everything that takes place on the converted land, together with additional ethanol production and use, that should serve as the basis for calculating corn ethanol’s total life-cycle carbon emissions in the case of dLUC .
Although Fargione et al. rightly considered land conversion and associated carbon loss, they relied on prior LCA studies , which were based on corn from long-standing cornfield, to estimate everything else. In so doing, they failed to recognize that newly converted land is generally not as fertile as cornfield persisting in cultivation and that corn ethanol originating from low-fertility land would provide smaller carbon benefits than corn ethanol originating from long-standing cornfield. Accounting for the actual yield of the converted land , as demonstrated by Yang and Suh , could substantially increase the time it takes for the use of corn ethanol to repay the carbon debt created by the initial land conversion. Exiting iLUC studies calculate corn ethanol’s total carbon emissions in the same way as do previous dLUC studies by adding carbon loss from land conversion to the carbon footprint of corn ethanol. When exposed with the same consequential reasoning, however, the iLUC literature commits the same error as committed in previous dLUC studies. But for iLUC effect it is beyond the actual yield or fertility of the converted land; what and how new crops are produced following land conversion matters. To drive home, let us consider a simple, hypothetical example of iLUC. Suppose, in response to increasing ethanol demand, part of U.S. corn was diverted to ethanol production at the expense of reduced exports to China. Total U.S. corn production and areas thus remained unchanged. This drove up Chinese corn prices and subsequently led Chinese people in rural areas to eat more rice, which drove up rice prices there and led Chinese subsistence farmers to convert reforested land to rice cultivation. What are the carbon consequences of corn ethanol expansion in this example? However, because U.S. corn production did not change or was not affected in this example, it is irrelevant to corn ethanol’s carbon consequences, as is the corn from longstanding cornfield in the case of dLUC. There was no additional carbon uptake from corn growth, nor were there additional carbon emissions from the use of agricultural inputs in corn production. What matters, instead, is the additional rice cultivation in China – which took place to compensate for the U.S. corn diverted to ethanol production – and associated carbon uptake and emissions . Of course, this is an extremely simplified example. Real-world consequences of U.S. corn ethanol expansion could be much more complicated, involving conversion of assorted natural habitats and different croplands brought into production in different countries. In any case, carbon uptake and emissions associated with whatever cropland being brought into production worldwide – including, likely, additional corn – should count towards the carbon consequences of ethanol expansion. Simply adding carbon loss from indirect land conversion across the world to the carbon footprint of U.S. corn ethanol is not meaningful from both theoretical and empirical perspectives. In addition to estimation of carbon loss from indirect land conversion , future studies need also direct efforts to account for what and how crops would be grown following land conversion and associated carbon uptake and emissions. In the chapter on carbon payback time , we assumed a perfect 1:1 displacement ratio between corn ethanol and gasoline on an energy basis, an assumption also used in previous carbon payback time studies .