What are the expected relative GHG emission savings of biofuels: uncertainty

The major sources of these uncertainties are mentioned in the RFA & DECC (2009) report, but they are not explained in detail there. The most comprehensive paper I have found so far which explains the uncertainties of GHG-saving potential of biofuels is the Johnson et al. (2011) article. You can explore the article to find out the details of the uncertainty sources, but here I will just review the points I found to be the most important and most interesting.

• One source of variability between models lies within which factors the model incorporates e.g. the modeller has to choose whether to include indirect emissions, such as due to land-use change and whether to stop at 2^nd or 3^rd order emission effects. I found this point most interesting, as the solution to avoiding this issue appears so straight-forward –  it seems that all one needs to do is incorporate as many factors as is possible, as surely this will make the model more accurate. Yet, world-class researchers seem to find this a difficult task!

I personally think that the best illustration of this is the Crutzen et al. (2008) study, which calculated that previous reports underestimated the climate change contribution of GHG emissions from biofuels by excluding the nitrous oxide emissions effects during cultivation. This has severe implications for policy-making: for example, rapeseed, a relatively common biofuel material at present, now potentially apparently contributes 1-1.5 times more towards global warming than fossil fuels instead of having a lower global warming impact as previously calculated. However, it seems that Crutzen et al. (2008) have not learnt from their own discovery of the importance of cultivation emissions, as their study excludes the emissions from fossil fuels used in the growing of biofuels. So why does this lack of incorporation of these variables continue happening?

At the moment, the best answer I could find was deduced from the Johnson et al. (2011) article, which is that the reason lies within the time and resources needed to account for all the possible factors that may affect GHG emissions of biofuel production. Ekval and Weidema point out that trying to include all the possible emission sources can continue indefinitely (Johnson et al., 2011). Johnson et al. suggest only including the most likely and the most significant parameters to account for this problem. However, I hope I have illustrated that this is not always so straight-forward at the moment, as it is often difficult to predict which factors are important without exploring all of them in the first place. This suggests that since this is impractical to do and all the possible sources of emissions have probably not been incorporated into current models yet, the present estimates of biofuel impacts may still change in the future.

• Another major source of uncertainty is the variation in time and space. It is impossible to decide ‘correctly’ on what data would constitute typical emissions on a larger temporal and spatial scale, as this would be ignoring differences in agricultural practices, in energy input and yield output and in soil carbon storage, as well as numerous other factors (Johnson et al., 2011). It is also impossible to predict accurately the future development in technology, agricultural practices and social and economic development. This will thus affect the use of biofuel and the emissions estimates from biofuel production.

• Uncertainty in allocating GHG emissions also arises where a material has multiple products. For example, growing corn for biofuel produces both grain and stover, where one of these products can be utilized in biofuel synthesis and the other used for another purpose, such as food. It would be near to impossible to separate the emissions due to biofuel alone here.

• The other sources of uncertainty lye within, for example, a lack of sufficiently detailed data on certain processes.

• Scenario uncertainty also exists, such as the percentage of biofuel to be used in transport in the future, which is subject to economic and political factors that can not always be predicted, for example.

In conclusion, there is a number of reasons for the often high variability in the estimates of the relative biofuel GHG savings between and within reports. This variability means it is important to keep in mind that the topic of biofuels being discussed in this blog is based on rather uncertain data. However, as with other such issues where full scientific certainty is impossible at present, management decisions still have to be made despite this uncertainty, so it is important to make the most well-informed judgement on the benefits of different actions possible. In order to be able to do this, knowing the potential sources of uncertainty in the data, some of which are outlined in this post, is useful.