This year is so far tied for the hottest year in a record dating back to 1850 in a new sign of a warming trend, the three major institutes which calculate global warming estimates told Reuters.
UN climate talks resume next week in CancĂșn, Mexico, where expectations are no longer for a comprehensive deal to slow warming, but smaller progress for example to curb deforestation, in a bid to agree a pact next year or later.
The previous conference in Copenhagen last year fell short of hopes, but about 140 countries have agreed a non-binding deal to try to limit warming to less than 2C above pre-industrial levels.
Temperatures are now about 0.8C above pre-industrial levels, and 2010 is about 0.5C above the 1961-1990 average, near the record, with two months' data still to collect.
Even with a possible cool end to the year, 2010 is expected to be no lower than third in a record where 1998 and 2005 are warmest. The UN panel of climate scientists says higher temperatures mean more floods, heatwaves and rising sea levels.
"I think it's too close to call. Based on these numbers it'll be second, but it depends on how warm November and December are," said Dr Phil Jones, director of the Climatic Research Unit (CRU), at the University of East Anglia, which says 1998 was the record year so far.
By contrast, scientists at Nasa say surface temperatures through October were above the previous record year, which it says was 2005. Differences between years are only a few hundredths of a degree.
"I would not be surprised if most or all groups found that 2010 was tied for the warmest year," said Nasa's Dr James Hansen.
And the US National Climatic Data Center (NCDC) at the National Oceanic and Atmospheric Administration, said 2010 was a "dead-heat" for the record. "Our data show 2010 being virtually tied with 1998, through October," said Deke Arndt, from NCDC.
The three institutes use similar observations, but in slightly different ways. For example, Nasa's Goddard Institute for Space Studies takes greater account of Arctic weather stations, where warming has been fastest.
Some sceptics have argued that because the last temperature peak was in 2005 or 1998, that global warming must have stalled.
Most scientists reject that view, saying that whether or not 2010 is the hottest year is less important than the long-term trend, which is up, due to human-caused greenhouse gas emissions. The period 2000-2009 was the warmest decade on record.
Scientists also point to natural variation, and in particular the El Nino Pacific weather phenomenon associated with warm weather worldwide. 1998 was a strong El Nino year.
"The trend is overwhelming, particularly over the past 50 years," said Rajendra Pachauri, head of the UN panel of climate scientists. "I wouldn't read these numbers for a particular year as very compelling, we have to take a historical view," he told Reuters.
In one of the biggest bets on climate change, James Annan, a climate scientist at the Frontier Research Centre for Global Change in Japan, has a $10,000 wager made in 2005 with two Russian solar physicists who are sceptical about global warming.
He will win if average world temperatures are higher from 2012-17 than they were from 1998-2003. "Things are progressing smoothly," he said.
The UN's World Meteorological Organisation will publish an estimate on 2 December of where 2010 ranks. It compiles data from a wider range of sources, both measured temperatures and climate models. It lists 1998 and 2005 as the warmest years. "We have indications that it would match one of the three warmest years," said Omar Baddour, head of climate data management operations in Geneva.