We welcome the October publication of the draft Water and food security report by The High Level Panel of Experts on Food Security and Nutrition and take particular note of the sections in the draft relevant to biofuels, which we summarize below:
While some kinds of renewable energy, such as wind and solar photovoltaic (PV) power, do not consume much water, other renewable energy processes such as concentrated solar power and biofuels consume significant quantities of water. . . . Biofuels can add pressure to ‘water supply and water quality problems’ (HLPE 2013) especially if irrigated (Lundqvist, de Fraiture and Molden 2008). Although regional variation is large, de Fraiture et al (2008) estimate that on average it takes around 2 500 litres of crop evapotranspiration and 820 litres of water withdrawals to provide one litre of biofuel. It is at the country or local level that the trade offs between water for food and water for biofuels are felt. For example, in India water for biofuels can compete directly with water for food such as cereals and vegetables (ibid). Also as concluded by the HLPE, biofuel production usually does not benefit small-holder farmers in water scarce contexts (HLPE, 2011).
We note that only one source is used for the draft's conclusions about the biofuels and water nexus, a desk research paper submitted in 2007 about China and India (de Fraiture, C., Giordano, M. And Liao, Y. 2008. Biofuels and implications for agricultural water use: blue impacts of green energy, Water Policy 10 Supplement 1: 67-81), which we will call the "Paper". We have no criticism of the Paper save one- which is that it seems unaware that most biofuels processes result in just as much animal feed as biofuel. Yet, in the draft Water and food security the Paper is (i) misquoted, (ii) unsuited as a source for global conclusions, and (iii) not actually about biofuels as all. We urge you to make changes to the final report to avoid misinterpretation and explain our reasoning below.
For almost a decade the Paper has been used by anti-biofuel campaigners to suggest that biofuels production facilities consume and pollute huge amounts of water, which, of course, the Paper does not say at all. In fact, where the Paper says "biofuels" it actually only means "biofuel feedstock". A modern maize ethanol plant pollutes no water and uses (often recycling this water) only 3 liters of water to produce one liter of ethanol.
Water and food security currently states both that "biofuels consume considerable amounts of water" and "on average it takes around 2 500 litres of crop evapotranspiration and 820 litres of water withdrawals to provide one litre of biofuel". Yet, the Paper takes pains to stress that US maize ethanol and EU rapeseed biodiesel are rain fed and have almost no water scarcity impacts, both directly and in context (e.g. "From a water perspective, it makes a large difference whether biofuel is derived from fully irrigated sugarcane grown in semi-arid areas or rain-fed maize grown in water-abundant regions."). To a lesser degree, the Paper makes the same point about Brazilian sugarcane. And, the Paper unfortunately ignores the undeniable fact that both maize ethanol and rapeseed biodiesel processes (as opposed to sugarcane ethanol) result in just as much high protein animal feed as biofuel, belying the suggestion that water used by the underlying crops should be booked only as an energy use.
We must take great exception to the suggestion that biofuel production "usually does not benefit small holder farmers", although we appreciate the context of the phrase that follows. Our concern is because most of the biofuel feedstock processed into biofuels in the Northern Hemisphere (whether ethanol or biodiesel) is produced by family farmers.
The "average" in 2006 is not the average in 2014. Almost no biofuel produced in the world's three largest biofuel producing areas (Brazil, Europe and the United States) is accurately described by the above statements in the current draft. Since these biofuels constitute the vast majority of 2014 biofuels, there is no basis to include either phrase above in the final report.
The Paper is about theoretical impacts in China and India, which is clearly what Table 1 of the Paper states. Accordingly, we respectfully request that the relevant passages of the draft be revised as follows:
While some kinds of renewable energy, such as wind and solar photovoltaic (PV) power, do not consume much water, other renewable energy processes may [such as concentrated solar power and biofuels[ consume significant quantities of water. . . . Biofuel feedstocks can add pressure to ‘water supply and water quality problems’ (HLPE 2013) especially if irrigated (Lundqvist, de Fraiture and Molden 2008). [Although regional variation is large, de Fraiture et al (2008) estimate that on average it takes around 2 500 litres of crop evapotranspiration and 820 litres of water withdrawals to provide one litre of biofuel.] It is at the country or local level that the trade offs between water for food and water for biofuels are felt, particularly in water-scare regions. For example, in India water for biofuels can compete directly with water for food such as cereals and vegetables (de Fraiture et al (2008)ibid). Also as concluded by the HLPE, biofuel production usually does not benefit small-holder farmers in water scarce contexts in the Southern Hemisphere (HLPE, 2011).
Respectfully yours,
Eric W. Sievers
CEO
Ethanol Europe Renewables Limited
Eric Sievers