By "very likely", the IPCC means greater than 90% probability.  The scientific body, in a report released in Paris today, forecasts temperatures will probably rise by between 1.8-4C (3.2-7.2F) by 2100.

But another study released on the eve of publication suggests its previous reports may have been too conservative.

Dr Rajendra Pachauri, the IPCC chairman, said:

"The science has moved on from what was possible in the Third Assessment report.

"If you see the extent to which human activities are influencing the climate system, the options for mitigating greenhouse gas emissions appear in a different light, because you can see what the costs of inaction are," he told delegates at the launch of the Fourth Assessment report.

The document released today by the IPCC is intended to be the definitive summary of climate change science. The agency has said it would use stronger language to assess humanity's influence on climatic change than it had previously done.

In 2001, it said that it was "likely" that human activities lay behind the trends observed at various parts of the planet; "likely" in IPCC terminology means between 66% and 90% probability.

Now, the panel concludes, it is at least 90% certain that human emissions of greenhouse gases rather than natural variations are warming the planet's surface.

"The understanding of [human] warming and cooling influences on climate has improved since the Third Assessment report, leading to very high confidence that the globally averaged net effect of human activities since 1750 has been one of warming," observed the scientists.

They conclude that temperatures will probably rise by between 1.8 C and 4C, though increases as small as 1.1C (2F) or as large as 6.4C (11.5F) are possible.

In 2001, using different methodology, the numbers were 1.4 (2.5F) and 5.8C (10.4F).  On sea level, there has been a more fundamental debate.

Computer models of climate do not generally include water coming into the oceans as ice caps melt. So the IPCC had to decide whether to exclude this from its calculations, or to estimate the effect of a process which scientists do not understand well but which could have a big impact.

They have gone for the former, more conservative approach, projecting an average rise in sea levels globally of between 28 and 43cm. The 2001 report cited a range of nine to 88cm.

As for climate change influencing the intensity of tropical storms in some areas of the world, the IPCC concluded that it was likely - meaning a greater probability than 66% - that rising temperatures were a factor.

But a study published on the eve of the IPCC report suggested that the international body's previous reports may have actually been too conservative.

Writing in the journal Science, an international group of scientists concluded that temperatures and sea levels had been rising at or above the maximum rates proposed in the last report, which was published in 2001.

The paper compared the 2001 projections on temperature and sea level change report with what has actually happened.

The models had forecasted a temperature rise between about 0.15C-0.35C (0.27-0.63F) over this period. The actual rise of 0.33C (0.59F) was very close to the top of the IPCC's range.

A more dramatic picture emerged from the sea level comparison. The actual average level, measured by tide gauges and satellites, had risen faster than the intergovernmental panel of scientists predicted it would.

The IPCC's full climate science report will be released later in the year, as will other chapters looking at the probable impacts of climate change, options for adapting to those impacts, and possible routes to reducing emissions of greenhouse gases.