There are gaps in scientific understanding making predicting the extent of climate change and sea level rises impossible.

That's the claim of Britain's highest scientific authority, the Royal Society.

The society's revised Guide to the Science of Climate Change has been interpreted as a retreat from politics by an organisation regarded as the world's most authoritative scientific body following the scandal that engulfed the UN's Intergovernmental Panel on Climate Change.

The society's new guide does not dismiss climate change or the need for co-ordinated global action to combat it.

However, it undercuts many of the claims of looming ecological disaster that have been made in a bid to gain public support for political action.

The opposition seized on the Royal Society's shift to demand Julia Gillard accept that views on climate change differ.

Opposition climate spokesman Greg Hunt said: "This is a reminder on why Julia Gillard is wrong to vilify people who have the audacity to disagree with her views.

"We respect the right of individuals to make up their own minds based on their own assessments."

Climate Change Minister Greg Combet said the Royal Society's switch would not have any influence on the government's push to put a price on carbon.

"The government accepts the climate science," Mr Combet said.

"The debate has moved on.

"We must now get on with the job of reducing carbon pollution and reforming our economy."

The society's report was written by a panel of prominent scientists chaired by professor John Pethica.

The reworking was in response to pressure from 43 fellows who argued the society had gone too far.

Ian Plimer, professor of mining geology at Adelaide University, said the society's statement was a "wonderful breath of honesty and fresh air from an organisation that has been politicised".

"Science is always uncertain," Professor Plimer said.

"Science doesn't work by voting.

"It is not a democracy, it works on evidence."

Despite the uncertainties, the Royal Society concludes that there is strong evidence that changes in greenhouse gas concentrations due to human activity are the dominant cause of global warming over the past half-century.

"It is not possible to determine exactly how much the Earth will warm or exactly how the climate will change in the future, but careful estimates of potential changes and associated uncertainties have been made," Professor Plimer said.

"Uncertainty can work both ways, since the changes and their impacts may be either smaller or larger than those projected."

On sea level change, the society said it was likely that, for many centuries, the rate of global sea-level rise would be at least as large as the rate of 20cm per century that has been observed.

However, it said there was insufficient understanding of the melting of the ice sheets in Greenland and West Antarctica to predict how much the sea level will rise above that observed in the past century.

The society's cautious approach is in contrast to the UN's 2007 IPCC report.

(The report said that "many millions more people are projected to be flooded every year due to sea-level rise by the 2080s").

Also to predictions by former US president Al Gore in his documentary An Inconvenient Truth that the collapse of a major ice sheet in Greenland or West Antarctica could raise global sea levels by six metres, flooding coasts and creating 100 million refugees.

The Royal Society says the greatest gap in understanding is being able to accurately model clouds and their impact on reflecting heat.

The strength of the uptake of CO2 by the land and oceans, which take up about half the emissions from human activity, is poorly understood, it says.

And there is little confidence in specific projections of future regional climate change, except at continental scales.

The Royal Society says that high-performance computers are expected to improve confidence in regional predictions.

There is also a possibility that unknown aspects of climate change could emerge and lead to significant modifications in our understanding.