The water footprint of bioenergy from Jatropha curcas L

The water footprint (WF) of energy crops should be calculated by relating the energy yield of a crop to its actual water use under actual climatic conditions during the growing season. Recently, average precipitation and additional irrigation to satisfy potential crop water requirements were used as input to the WF computation (1). We disagree with this approach because it relates optimized crop water use to actual yields, thereby introducing bias towards inefficient WF for low-yielding crops under suboptimal rain-fed conditions.

In particular, the presented WF of the pan-tropical perennial species Jatropha curcas L. deserves correction. A too simplistic computation discards this multi-purpose tree as an inefficient energy source, whereas it may provide a viable alternative in semi-arid areas, contributing to increased resource use efficiency, erosion control, crop diversification, local energy supply, and rural development. Jatropha has a range of characteristics for coping with water stress. Its physical constitution changes, and it is suggested to combine C3-/CAM photosynthesis in succulent stems (water storage), with leaves shifting from C3-metabolism to more water-efficient CAM (2). Low water use is supported by field data (3), but claims that combine this trait with high production potentials are unrealistic (4).

Our objections to the presented WF for jatropha concern the use of low-yield data (from nonirrigated young plantations) connected to simulated water use of mature plantations under optimal growing conditions, without considering specific characteristics of soils and species. Background information (5) did not supply essential information for sound WF calculations with appropriate values for jatropha crop factors to correct reference evaporation.

Furthermore, the remaining press cake of jatropha seeds after oil extraction should be considered for energy production. Except for some Central American accessions, all plant parts are toxic and unlike other oil species, its press cake cannot be used for animal feed without detoxification. Roughly, jatropha seeds provide 1/3 oil and 2/3 press cake with higher heating values of 37.7 MJ/kg oil and 17.5 MJ/kg press cake. Considering the division of 33% and 67%, both fractions give similar energy production and double the energy yield, or halve the WF per GJ produced.

In South Africa, nonirrigated unfertilized 4-year-old jatropha (4.5 × 3.0 m; 741 trees/ha) yielded 1,286 kg/ha dry seeds in a growing season of 8.5 months with 652 mm rainfall. With 35% oil, this represents 450 kg/ha oil (or 489 L/ha oil at 0.92 kg/L) and 836 kg/ha press cake, delivering 31.6 GJ/ha. Over the growing season, our water-balance model simulated total transpiration and soil evaporation of 4,052 m³/ha, well in agreement with field observations (3). The concurrent WF is 8,281 L of water per L of oil and 128 m³ of water per GJ; not even 1/3 of the WF of soybean, comparable to the WF of cassava, and only 1/5 of the WF obtained by using potential crop water requirements (1).