Until recently most scientists have compared fossil fuels and biofuels by totting up their direct greenhouse-gas emissions. Direct emissions include things like the diesel used to drive the tractor to cultivate the maize, but not the carbon in the biofuel itself.

Comparing like for like, fossil fuels were considered to produce more carbon dioxide than biofuels, which has led to the assumption that fossil fuels cause more greenhouse warming than biofuels. However, when 'indirect' emissions and timing of emissions are taken into consideration, biofuels begin to lose their green credentials.

Michael O'Hare of the University of California, in Berkeley, US, and his colleagues developed a model to assess the greenhouse-gas impact from both fossil fuels and biofuels over their entire lifecycle. In each case they studied how carbon dioxide emissions change over time, and how long-lived the carbon dioxide is.

In a case study they used their Biofuels Time Integrated Model of Emissions (BTIME) to compare the production of maize ethanol with that of ordinary gasoline, over a 100-year period. For the maize ethanol there is a large release of carbon dioxide initially, when forests and pastures are converted to grow replacement food (food that would have been grown where the new biofuel crop is growing).

There is then a steady release of carbon dioxide (60 g MJ of energy) during the years of production. When production ceases, perhaps after 25 years, the cleared land begins to revert to its original state and a small amount of carbon dioxide is sequestered.

Gasoline, by contrast, does not produce an initial pulse, but instead a steady stream of carbon dioxide (94 g per MJ of energy) over the same 25-year period. "Fossil fuels emit the carbon as you use them, whereas crop-based biofuels emit an initial pulse of carbon dioxide that is large," O'Hare told environmentalresearchweb.

In a parallel project, they estimated the "indirect" discharges from worldwide land clearing, caused as economic signals move through world food markets when land is used for fuel instead of food. Putting the findings together from both models they were able to calculate the direct and indirect emissions associated with biofuels, and consider how the emissions change over time.

Once carbon dioxide gets into the atmosphere it takes a long time to leave. "It follows an exponential decay pattern, losing a constant percentage every year," explained O'Hare. The sudden and early nature of the emissions associated with biofuels makes them more damaging than fossil fuels.

In the case of maize ethanol it leads to a higher abundance of carbon dioxide in the atmosphere for the first 15 years of its production, compared to gasoline. Even when the gasoline emissions catch up with maize ethanol, the global-warming damage from the maize ethanol emissions is greater, because they have been in the atmosphere for longer. The results are published in the journal Environmental Research Letters.

Such calculations are crucial when it comes to climate policy. Many countries are now starting to consider carbon taxes and it is essential that the taxes reflect the damage that a product causes over similar timescales.

O'Hare stressed that it is only biofuels that are grown in locations where food might otherwise have been grown that compare unfavourably with fossil fuels. "Biofuels that don't compete with food for land are fine on this measure, such as ethanol that is produced from trash, or from ocean algae," he said.

And it is possible that some biofuels could become less damaging. "Increasing yields, by making ethanol out of the whole plant for example, could reduce their overall impact," said O'Hare.