Some want action now to curb climate-changing emissions, saying that will cost little today but more if we delay, while others urge a slower approach, saying uncontrolled climate change will cost little or nothing in the short-term.

The report by UK government scientist Nicholas Stern to the G8 nations and major emerging countries including China, India and Brazil, may favour the first argument, according to some sources who contributed to it.

Also in that camp is Nobel laureate and former World Bank chief economist Joseph Stiglitz.

"The net cost (of taking action) is even potentially negative but certainly is not very significant," said Stiglitz.

Financial benefits, or negative costs, occur from energy efficiency savings.

"We need to start doing something now," Stiglitz said. "There is a risk of very rapid climate change."

Robert Mendelsohn, professor at the Yale School of Forestry and Environmental Studies, argues that such negative costs may still be less than the benefits.

He sees a net global warming bonus in the near-term, as higher farming yields in northern countries offsets damage elsewhere, especially in Africa.

"In that sense it doesn't make sense to spend money right now," Mendelsohn said, adding that beyond 2050 and a 2 degrees celsius rise the damage and need for action grows.

He added that he does not cost species extinctions and health effects, and only crudely measures the cost of island inundations.

Richard Tol, Senior Research Officer at Ireland's Economic and Social Research Institute, has a similar stance.

"(My damage estimate) does hide some things that some people will get very upset about," Tol said. "From an economic perspective small island states are so tiny and people are moving out of there anyway."

As an example Tol estimates the welfare loss of the Maldives submerging at three times the inhabitants' annual salaries, in addition to the 100 percent loss of the country's GDP.

Citizens are happy to value the preservation of the global ecosystem at a cost of 50 euros per person per year, Tol says, but added he does not factor in the risk of rapid sea level rise.

UNDERESTIMATES

A third camp treats with suspicion both cost and damage estimates, fearing they could be huge underestimates.

"I worry that existing damage estimates have little to do with what we'll actually see," said John Reilly, senior research scientist at MIT. "They do not really value the widespread ecological changes that are likely to occur."

A particular concern is the cost of runaway climate change, where temperature rises spin out of control, and which could trigger knock-on disasters like conflicts, or sudden sea level rise which could wipe out part or whole countries like the Netherlands, Egypt and Bangladesh.

On the costs of policies, the concern is of over-optimistic assumptions about the update of new clean energy technologies, with recent oil price hikes, for example, spurring less adoption impact than some had expected.

The annual costs of tackling climate change escalate rapidly the tougher the action.

Reilly estimates the cost of staying within a 3 degrees temperature rise at 2 percent of global GDP in 2100, but at some 8 to 10 percent of GDP to stay under 2 degrees -- or some US$25 trillion in 2100 money -- seen as a danger threshold.

Mendelsohn estimated that the top end of possible temperature rises, nearly 6 degrees, would cost up to 2 percent of GDP in 2100.

Britain's Stern will present his findings to ministers in Mexico next week, a month before countries start talks -- expected to last years -- on a successor to take the Kyoto protocol beyond 2012.