Nasa officials have admitted that for the past seven years they have used the wrong temperature statistics to assess global warming trends.

The mistake was spotted by a blogger, and forced Nasa to declare that last year was only the fourth-hottest year on record in the United States, rather than the third hottest.

Climate scientists were insistent yesterday that the flawed data had a negligible effect on global warming statistics, and none at all on the overall warming trend.

The figures related to temperature readings in the US, which scientists at the Nasa Goddard Institute of Space Science had neglected to adjust to be compatible with other readings. The changes meant that, on average, Nasa’s figures on US temperatures since 2000 had been too high by up to 0.15C (0.27F).

The error in the data was seized upon by global warming sceptics, who regarded it as a perfect example of why climate change science should be doubted.

James Hansen, the director of the Goddard Institute, maintained that while the flawed data had skewed figures “noticeably” for the past seven years, the corrected figures left evidence of long-term warming intact.

Other scientists hit back at global warming sceptics by pointing out that the error applied to only 48 states of the US, with Alaska and Hawaii excluded, and was negligible in its impact.

Moreover, they argued, it had no impact at all on the rankings of the hottest years when looked at globally rather than in the US, which accounts for only 2 per cent of the Earth’s surface.

David Parker, of the Met Office Hadley Centre, said of the effect on global averages: “The effect is so small that you couldn’t see it on a graph. They were of the order of a thousandth of a degree. It really has no impact.”

Further adjustments were made to the US data this year to take into account new information and new interpretations of how it should be adjusted. This meant that 1934 took over from 1998 as the hottest year on record in the US, and 1921 moved into third place above 2006.

Reto Ruedy, of the Goddard Institute, said that the differences between 1934 and 1998 were so slight that it was likely that they would swap positions again as information is analysed anew.

Paul Simons, Times Weatherman, writes:

There is nothing unique to the US about this. We have had to make exactly the same adjustment in the UK. For instance, the weather station at Kew Gardens was once a rural idyll, but is now firmly planted in the heat of West London.

The temperature adjustments that have been made to the US data are almost insignificant for the global temperature records. The US adjustments amount to 0.15 per cent of yearly average temperatures over the past six years. But the US land surface is only 2 per cent of the world land area, so the effect on global records is negligible.

But global warming is real and it is definitely happening – the Earth’s temperature this July was the seventh warmest July on record, June was the third warmest and January and April were the second hottest for the first six months of the year across the globe.

Times Recommends

• Postmen: We bind the nation together
• Tewkesbury: when the memories flood back
• Cast Offs actress: I’m happy that my face fits