Bill Hare from the Potsdam Institute for Climate Impact Research in Germany was the coordinator for the section on sea level rise for the benchmark 2007 Assessment Report 4 (AR4) of the IPCC. He now says it is "likely that IPCC AR4 sea level rise projections are biased low".

New research carried out since AR4 shows that the "risk of additional sea level rise from both the Greenland and Antarctic ice sheets may be larger than projected and could occur on century time scales".

"Ice dynamical processes seen in recent observations but not fully included in (AR4) ice sheet models could increase the rate of ice loss," Hare said on the sidelines of the Dec 1-12 summit of the UN Framework Convention on Climate Change (UNFCCC) here.

Hare, considered the world's leading scientist on polar ice caps, delivered a dire warning to the over 9,000 delegates attending the Poznan summit, which is stuck over the ways to limit global warming by 2100 to two degrees Celsius above the levels of the pre-Industrial Age.

In the Greenland ice sheet that covers the north polar region "warming as low as 1.9 degrees Celsius above pre-industrial could lead to irreversible meltdown", Hare said, adding: "This could lead to a sea level rise of 6-7 metres."

In the west Antarctic ice sheet too, "risk of disintegration increases with warming", Hare said. "Observed losses (have) increased 75 percent (in the) last 10 years." This can lead to further rise of 4-5 metres in the sea level.

The really apocalyptic one is the main Antarctic ice sheet, whose melting would lead to a sea level rise of about 52 metres.

But even without that, the melting of the Greenland ice sheet alone would drown thousands of islands and most coastal cities around the world, with the Ganges delta among those most at risk.

Hare explained the science behind the accelerated melting of the Greenland ice cap and said the retreat since 1997 of the Jakobshavn Isbrae Glacier - the largest outlet glacier from Greenland - appeared to be due to warm ocean waters.

When asked what is the future of the Greenland ice sheet, Hare said: "Contraction (of ice sheet is) projected to continue; "Virtually complete loss with sea level rise of about seven metres if global average warming were sustained for millennia in excess of 1.9 to 4.6 degrees Celsius relative to pre-industrial values;

"Projected 21st (century) temperatures in Greenland comparable to last interglacial period 125,000 years ago."

What about Antarctica? According to Hare, there was virtually loss in east Antarctica, but in the west there were "widespread losses along the Bellingshausen and Amundsen seas and increased loss by 59% in 10 years".

In the Antarctic Peninsula, he added, "losses (are) increasing" but so far they were "concentrated along narrow channels occupied by outlet glaciers".

Hare concluded:

* Greenland ice sheet decay seems likely at or above 1.5 degrees Celsius warming. If present ice sheet behaviour continues then this threshold could be lower.

* West Antarctic Ice Sheet (WAIS) danger points appear to be in the range 2-4.5 degrees Celsius.

* Accelerating loss of ice from the Amundsen Sea Sector of the WAIS gives rise to very substantial concerns at lower temperatures.

* Uncertainty in sea level rise projections very large.

* Sea level rise likely much higher than IPCCAR4 because AR4 could not include all processes and current trends are larger than hose modelled.

* To keep sea level rise to 0.5 metres by 2100 will likely require emissions to be at lowest level assessed by IPCC AR4.

* Benefits of mitigation in the 21st century for sea level rise in the long term are high and are more certain than sea level rise projections.

Stefan Rahmstorf, another scientist from the Potsdam institute and also an IPCC expert, said that between 1961 and 2003, the melting of ice sheets had contributed 25 percent to global sea level rise, but since 2003 this had gone up to 40%.