While afforestation – planting trees – is allowed under the Kyoto Protocol as a way for developed nations to offset their carbon emissions, avoiding deforestation, i.e. leaving the trees standing, is not. Conservationists argue that it should be, pointing out that retaining existing forest has stronger benefits for biodiversity and other ecosystem services than planting new forest.

A financial mechanism for including 'avoided deforestation', known as REDD (Reducing Emissions from Deforestation and Degradation) is up for discussion at the next round of Kyoto talks, in Copenhagen in 2009. The new economic study adds to the case for it, using models of global land use to show that avoided deforestation compares favourably in terms of cost to other carbon emission abatement options.

The study was a collaboration between three teams who have independently devised economic models of global land use. The teams were from Ohio State University, US, Lawrence Berkeley National Laboratory, US, and the International Institute of Applied Systems Analysis in Austria. They set out to estimate how much carbon emission could be avoided from 2005 to 2030 by taking economically realistic opportunities not to cut down forest, and how much this would cost.

By using three different models to address the problem, the researchers encompassed in their answers the variability in estimates of forest cover, carbon storage in forests and costs of land management.

In each model, global land use develops according to assumptions about future trade, technology and world population. The models assume a financial mechanism that pays for keeping forest, and a constant carbon price, set between $0 and $100 per tonne (t) of CO2 for each model run. Increasing the price of carbon means more forest is retained, and the emissions avoided can be calculated for a given carbon price by comparing total emissions with emissions from the model run when carbon has no value (equivalent to no avoided deforestation mechanism).

The results show that avoiding deforestation can provide substantial emission reductions. When carbon costs $20 per t CO2 for example, global emissions are reduced by 1.6 – 4.3 Gt CO2/yr over the 25-year period. To put this in context, the latest assessments suggest society needs to reduce carbon emissions by 3.5 Gt CO2 /yr to stabilise CO2 levels in the atmosphere at 550 ppm.

The cost to nations of avoiding deforestation, in terms of incomes from land foregone, is relatively low. A 10% reduction in deforestation from 2005-2030 would cost $2-5 per t CO2, and a 50% reduction would cost $10-21 per tonne CO2. "These are well within the range of costs for other climate change policy options," says Brent Sohngen, one of the report’s key authors from Ohio State University. A recent US study estimated a cost of $9 per t CO2 for reducing emissions by altering patterns of energy use.

Carbon payments could provide a powerful incentive to reduce deforestation. If the carbon price was a modest $10 per t CO2, standing forests could generate an income of more than $250 per hectare per year for some landowners, according to these models. Devising a reliable mechanism to reward avoided deforestation may be tricky, but Sohngen is optimistic. "If you create an economic incentive to avoid deforestation," he says, "people will find a way to measure, monitor and verify it."

The researchers reported their findings in Proceedings of the National Academy of Sciences (PNAS).