A comprehensive new analysis of water use in biofuel crop production finds that jatropha, an oil-rich plant championed for its ability to grow in arid regions where food crops cannot, is the biggest water hog of them all.

Researchers from the University of Twente, in the Netherlands, report in a recent issue of the Proceedings of the National Academy of Sciences that jatropha requires five times as much water per unit of energy as sugarcane and corn, and nearly ten times as much as sugar beet–the most water-efficient biofuel crop, according to the same study.

In recent years, as corn and other biofuel came under fire for driving up the cost of food production, some biofuel producers turned to Jatropha curcas, a weed that grows wild throughout the tropics and semitropics and produces seeds rich in oil.

In 2007, the oil-industry heavyweight BP teamed up with British biofuels company D1 Oils on a five-year, £80 million project to cultivate the plant in India, Southeast Asia, and Southern Africa. Together, the companies have planted more than 200,000 hectares so far. And the plant made headlines again late last year, when it became the first non-food-based biofuel to power a jet engine. But mounting evidence suggests that jatropha is not as ideal as once thought.

“The claim that jatropha doesn’t compete for water and land with food crops is complete nonsense,” says study coauthor Arjen Hoekstra. The researcher says it’s true that the plant can grow with little water and can survive through periods of drought, but to flourish, it needs good growing conditions just like any other plant. “If there isn’t sufficient water, you get a low amount of oil production,” Hoekstra says.

Hoekstra and his colleagues assessed the water footprint of 13 different biofuel crops. Their calculations included regional estimates of how much rainwater each crop received and how much additional water would be required through irrigation for optimal growth. The study also considered evaporation rates during the growing season in the main production areas of each crop, and the average yields of each from 1997 to 2001. The figures were then averaged by country and globally to come up with a single water-footprint figure–per liter of ethanol or biodiesel–for each crop.

“You see a big difference depending on the country where the biomass is produced, different climates, different agricultural practices, the crop being used, whether it is a starch or sugar crop used for bioethanol, an oil crop for biodiesel, or a crop that is burned for electricity generation,” Hoekstra says.

The team calculated that jatropha requires an average of 20,000 liters of water for every liter of biodiesel produced in India, Indonesia, Nicaragua, Brazil, and Guatemala–the only countries for which jatropha production figures were available. For all the other crops, the researchers used much more comprehensive–and thus truly global–data from the Food and Agriculture Organization of the United Nations. Soybeans and rapeseed, the two other biodiesel crops considered in the study, were next highest in terms of water consumption, each requiring roughly 14,000 liters of water per liter of fuel.

Chris Somerville, director of the Energy Biosciences Institute at the University of California, Berkeley, says he’s not that surprised by the study’s findings. “Jatropha is a wild species and probably has a terrible harvest index [ratio of yield to the total harvest] because no breeding has been done yet,” he says.

Somerville says that interest in jatropha was driven largely by European Union (EU) mandates for biodiesel production that were reduced in December 2008 due to environmental concerns that biofuels–in particular, biodiesel from palm oil–were causing the destruction of rain forests and wetlands. “I don’t know if we’ll see the death of jatropha, but we certainly see a lot less demand for it in Europe now than a couple of years ago, when there was a real scramble for it,” Somerville says.

Another recent study, carried out by Friends of the Earth, found that jatropha plantations in Swaziland run by BP and D1 Oils were taking land and water away from food crops in a country already suffering from chronic food shortages.

Somerville says that jatropha and other biodiesel crops will likely be pushed out by much higher yields of cellulosic ethanol in developed countries in the coming decade, but that the plants may continue to fill a niche. “The developing world may continue to see a big demand for jatropha and other vegetable oils because capital investment is much less than for ethanol and especially the highly technical processes of cellulosic fuels,” he says.

Henk Joos, who is the plant science director at D1 Oils, contends that the EU mandates still call for large quantities of biodiesel and says that newer, higher-yield strains of jatropha could solve many of the plant’s water-use issues. Joos and his team are crossbreeding different strains of jatropha to increase seed production and to maximize the seeds’ oil content, and they’re developing processes that allow the remaining seed biomass to be used for animal feed.

In 2006, the Energy and Resources Institute (TERI), an Indian research group, began a 10-year, $9.4 million effort to develop jatropha that included genetically engineering seeds to have higher oil content. Nibhi Chanana of TERI says that the group is still three to four years away from isolating the genes that control for oil production.